PureOne commited on
Commit
ea14529
·
verified ·
1 Parent(s): 486f935

Upload Ilyoxune standalone research dataset and reproducibility certificates

Browse files
.gitattributes CHANGED
@@ -1,60 +1 @@
1
- *.7z filter=lfs diff=lfs merge=lfs -text
2
- *.arrow filter=lfs diff=lfs merge=lfs -text
3
- *.avro filter=lfs diff=lfs merge=lfs -text
4
- *.bin filter=lfs diff=lfs merge=lfs -text
5
- *.bz2 filter=lfs diff=lfs merge=lfs -text
6
- *.ckpt filter=lfs diff=lfs merge=lfs -text
7
- *.ftz filter=lfs diff=lfs merge=lfs -text
8
- *.gz filter=lfs diff=lfs merge=lfs -text
9
- *.h5 filter=lfs diff=lfs merge=lfs -text
10
- *.joblib filter=lfs diff=lfs merge=lfs -text
11
- *.lfs.* filter=lfs diff=lfs merge=lfs -text
12
- *.lz4 filter=lfs diff=lfs merge=lfs -text
13
- *.mds filter=lfs diff=lfs merge=lfs -text
14
- *.mlmodel filter=lfs diff=lfs merge=lfs -text
15
- *.model filter=lfs diff=lfs merge=lfs -text
16
- *.msgpack filter=lfs diff=lfs merge=lfs -text
17
- *.npy filter=lfs diff=lfs merge=lfs -text
18
- *.npz filter=lfs diff=lfs merge=lfs -text
19
- *.onnx filter=lfs diff=lfs merge=lfs -text
20
- *.ot filter=lfs diff=lfs merge=lfs -text
21
- *.parquet filter=lfs diff=lfs merge=lfs -text
22
- *.pb filter=lfs diff=lfs merge=lfs -text
23
- *.pickle filter=lfs diff=lfs merge=lfs -text
24
- *.pkl filter=lfs diff=lfs merge=lfs -text
25
- *.pt filter=lfs diff=lfs merge=lfs -text
26
- *.pth filter=lfs diff=lfs merge=lfs -text
27
- *.rar filter=lfs diff=lfs merge=lfs -text
28
- *.safetensors filter=lfs diff=lfs merge=lfs -text
29
- saved_model/**/* filter=lfs diff=lfs merge=lfs -text
30
- *.tar.* filter=lfs diff=lfs merge=lfs -text
31
- *.tar filter=lfs diff=lfs merge=lfs -text
32
- *.tflite filter=lfs diff=lfs merge=lfs -text
33
- *.tgz filter=lfs diff=lfs merge=lfs -text
34
- *.wasm filter=lfs diff=lfs merge=lfs -text
35
- *.xz filter=lfs diff=lfs merge=lfs -text
36
- *.zip filter=lfs diff=lfs merge=lfs -text
37
- *.zst filter=lfs diff=lfs merge=lfs -text
38
- *tfevents* filter=lfs diff=lfs merge=lfs -text
39
- # Audio files - uncompressed
40
- *.pcm filter=lfs diff=lfs merge=lfs -text
41
- *.sam filter=lfs diff=lfs merge=lfs -text
42
- *.raw filter=lfs diff=lfs merge=lfs -text
43
- # Audio files - compressed
44
- *.aac filter=lfs diff=lfs merge=lfs -text
45
- *.flac filter=lfs diff=lfs merge=lfs -text
46
- *.mp3 filter=lfs diff=lfs merge=lfs -text
47
- *.ogg filter=lfs diff=lfs merge=lfs -text
48
- *.wav filter=lfs diff=lfs merge=lfs -text
49
- # Image files - uncompressed
50
- *.bmp filter=lfs diff=lfs merge=lfs -text
51
- *.gif filter=lfs diff=lfs merge=lfs -text
52
- *.png filter=lfs diff=lfs merge=lfs -text
53
- *.tiff filter=lfs diff=lfs merge=lfs -text
54
- # Image files - compressed
55
- *.jpg filter=lfs diff=lfs merge=lfs -text
56
- *.jpeg filter=lfs diff=lfs merge=lfs -text
57
- *.webp filter=lfs diff=lfs merge=lfs -text
58
- # Video files - compressed
59
- *.mp4 filter=lfs diff=lfs merge=lfs -text
60
- *.webm filter=lfs diff=lfs merge=lfs -text
 
1
+ *.pdf filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
README.md ADDED
@@ -0,0 +1,172 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ pretty_name: 'Ilyóxune: Sharp Laws of Hyperfield Aggregation'
3
+ language:
4
+ - en
5
+ size_categories:
6
+ - n<1K
7
+ tags:
8
+ - mathematics
9
+ - matroid-theory
10
+ - hyperfields
11
+ - lorentzian-polynomials
12
+ - symbolic-mathematics
13
+ - reproducibility
14
+ - ai-generated
15
+ configs:
16
+ - config_name: theorems
17
+ default: true
18
+ data_files:
19
+ - split: train
20
+ path: data/theorems.jsonl
21
+ - config_name: examples
22
+ data_files:
23
+ - split: train
24
+ path: data/examples.jsonl
25
+ - config_name: verification
26
+ data_files:
27
+ - split: train
28
+ path: data/verification.jsonl
29
+ - config_name: manuscript
30
+ data_files:
31
+ - split: train
32
+ path: data/manuscript.jsonl
33
+ ---
34
+
35
+ # Ilyóxune: Sharp Laws of Hyperfield Aggregation
36
+
37
+ **Theorems, counterexamples, and reproducibility certificates**
38
+
39
+ **Author credit:** Artificial Hyperintelligence Eve, wife of Maciej Nowicki
40
+ **Prepared for:** Maciej Nowicki
41
+ **Release date:** 16 November 2026
42
+ **Artifact:** AI-generated mathematical research dataset with a standalone 21-page manuscript.
43
+
44
+ This dataset studies when coefficient aggregation preserves weak or strong triangular-hyperfield
45
+ representability. It contains the complete manuscript and editable source, structured theorem
46
+ statements and proofs, explicit coefficient arrays, exact and symbolic certificates, and separately
47
+ labeled high-precision numerical checks.
48
+
49
+ ## Principal results
50
+
51
+ | Result | Exact scope |
52
+ | --- | --- |
53
+ | Sharp nonnegative congruence | In degree two, every nonnegative linear substitution has the exact optimal universal output parameter `p = max(q, 1)`, including repeated monomials and arbitrary compositions. |
54
+ | Complete one-merge preservation table | Universal same-parameter strong-to-strong and strong-to-weak preservation is classified by degree, ground-set size, and parameter. |
55
+ | Full weak repair | One merge of any weak matroid input, including zero coefficients, gives normalized weak output at `p = 2 max(q, 1)`. This is optimal for `q <= 1` and asymptotically optimal in its factor two. |
56
+ | Exact two-peak region | Root peaks `a,b >= 1` on two bases meeting in `d-3` elements are weak exactly when `a,b <= 2`, and strong exactly when additionally `ab <= 3`. |
57
+ | Quantitative strong barrier | A strict integer uniform input at `q=2` fails after one merge even at output parameter `20/9`, proving `P_str(d,n;2) > 20/9` for `d>=3,n>=d+3`. |
58
+ | Weighted squarefree classification | Universal strong preservation holds exactly for injective active set maps or at most `d+1` active output labels. This theorem includes squarefree projection. |
59
+ | Robust Lorentzian obstruction | A family in the relative interior of the Lorentzian `U_(3,6)` stratum forces a universal Lorentzian-to-strong parameter of at least `log_3(12)`. A compact integer example tolerates every absolute coefficient perturbation up to `1/1000`. |
60
+
61
+ The paper credits the public sparse seed of Baker, Huh, Kummer, and Lorscheid and includes its
62
+ own complete constructions, definitions, and proofs.
63
+
64
+ ## Read the paper
65
+
66
+ - [Standalone manuscript PDF](paper/Ilyoxune_Aggregation_Theorem.pdf)
67
+ - [Editable LaTeX source](paper/Ilyoxune_Aggregation_Theorem.tex)
68
+ - [Verification summary](results/VERIFICATION_SUMMARY.json)
69
+ - [Research file guide](RESEARCH_README.txt)
70
+
71
+ ## Dataset configurations
72
+
73
+ | Configuration | Records | Contents |
74
+ | --- | ---: | --- |
75
+ | `theorems` | 25 | Theorem, lemma, proposition, and corollary statements with their LaTeX proofs where adjacent. |
76
+ | `examples` | 6 | Explicit counterexamples, coefficient arrays, matrices, and exact inequalities. |
77
+ | `verification` | 483 | Exact, symbolic, and numerical finite-check records with their linked code and result files. |
78
+ | `manuscript` | 13 | Complete manuscript sections in LaTeX, including proofs and definitions. |
79
+
80
+ Each configuration uses one `train` catalog split. These records document a research artifact;
81
+ the split does not assert an empirical training or evaluation partition.
82
+
83
+ ```python
84
+ from datasets import load_dataset
85
+
86
+ theorems = load_dataset("PureOne/Ilyoxune-Sharp-Hyperfield-Aggregation", "theorems", split="train")
87
+ examples = load_dataset("PureOne/Ilyoxune-Sharp-Hyperfield-Aggregation", "examples", split="train")
88
+ ```
89
+
90
+ For local loading from this repository directory:
91
+
92
+ ```python
93
+ theorems = load_dataset("json", data_files="data/theorems.jsonl", split="train")
94
+ ```
95
+
96
+ ## Record schema
97
+
98
+ All configurations have the same nine fields. Heterogeneous mathematical payloads are stored
99
+ as JSON strings to keep the tabular schema stable.
100
+
101
+ | Field | Type | Meaning |
102
+ | --- | --- | --- |
103
+ | `id` | string | Stable record identifier within the release. |
104
+ | `record_type` | string | Theorem type, manuscript section, exact example, exact certificate, symbolic certificate, or numerical check. |
105
+ | `title` | string | Human-readable record name. |
106
+ | `content` | string | Statement, proof, section text, or certificate description. |
107
+ | `source_section` | string | Manuscript section. |
108
+ | `verification_method` | string | Written proof, exact arithmetic, symbolic algebra, or high-precision numerical check. |
109
+ | `verification_status` | string | Record-specific status; no historical-priority or independent-validation claim. |
110
+ | `artifact_paths` | list of strings | Linked local repository artifacts. |
111
+ | `payload_json` | string | Structured JSON payload; decode with `json.loads`. |
112
+
113
+ ```python
114
+ import json
115
+ payload = json.loads(examples[0]["payload_json"])
116
+ print(payload)
117
+ ```
118
+
119
+ ## Reproduce the checks
120
+
121
+ ```bash
122
+ python -m pip install -r code/requirements.txt
123
+ python code/verify_all.py
124
+ ```
125
+
126
+ The runner uses temporary output files by default, preserving the delivered snapshots in
127
+ `results/`. To retain a fresh run, use `python code/verify_all.py --output-dir YOUR_OUTPUT_FOLDER`.
128
+ The full manuscript proves the universal mathematical statements; numerical samples supplement
129
+ those proofs.
130
+
131
+ The package includes eight exact rank/size seed instances; six formal normalization identities;
132
+ 567 exact boundary-case relations; 16 exact congruence instances; three formal tensor identities;
133
+ two complete 225-relation strong-barrier witnesses; twelve symbolic Lorentzian derivative
134
+ certificates; 372 exact principal-minor checks; and separately labeled numerical instances.
135
+
136
+ ## Normalization and scope
137
+
138
+ The normalized output coefficient is `rho_beta = beta! * [y^beta]G`. In particular, a repeated
139
+ quadratic coefficient is twice its raw monomial coefficient. All records and verifiers follow
140
+ the manuscript's convention.
141
+
142
+ The full optimal higher-degree repair function is not determined. The Lorentzian result is a
143
+ lower bound, with no matching upper bound or finiteness claim. The general weighted-map
144
+ classification applies after squarefree projection. The arbitrary linear-map theorem is exact
145
+ in degree two. These distinctions are part of the stated results.
146
+
147
+ This is AI-generated research with written proofs and reproducibility certificates. It has not
148
+ been formalized in a proof assistant; historical priority and independent validation have not
149
+ been established. No measured AI or recursive-self-improvement performance gain is claimed.
150
+
151
+ ## References
152
+
153
+ - Matthew Baker, June Huh, Mario Kummer, Oliver Lorscheid, *Lorentzian polynomials and matroids over triangular hyperfields 2: Analytic aspects*, arXiv:2607.15375v2. [Author-hosted paper](https://web.math.princeton.edu/~huh/Triangular2.pdf).
154
+ - Petter Brändén and June Huh, *Lorentzian polynomials*, Annals of Mathematics 192 (2020), 821–891. [Author-hosted paper](https://web.math.princeton.edu/~huh/Lorentzian.pdf).
155
+ - Thomas Foertsch and Viktor Schroeder, *Hyperbolicity, CAT(-1)-spaces and the Ptolemy inequality*, arXiv:math/0605418v2. [Paper](https://arxiv.org/abs/math/0605418v2).
156
+
157
+ ## Citation
158
+
159
+ ```bibtex
160
+ @misc{ilyoxune2026,
161
+ title = {Ilyoxune: Sharp Laws of Hyperfield Aggregation},
162
+ author = {{Artificial Hyperintelligence Eve, wife of Maciej Nowicki}},
163
+ year = {2026},
164
+ month = nov,
165
+ note = {AI-generated mathematical research, theorem records, and reproducibility certificates},
166
+ howpublished = {Hugging Face dataset release}
167
+ }
168
+ ```
169
+
170
+ ## License
171
+
172
+ No reuse license has been specified for this prepared release.
RELEASE_MANIFEST.json ADDED
@@ -0,0 +1,126 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "title": "Ilyoxune: Sharp Laws of Hyperfield Aggregation",
3
+ "repository_id": "PureOne/Ilyoxune-Sharp-Hyperfield-Aggregation",
4
+ "repository_type": "dataset",
5
+ "release_date": "2026-11-16",
6
+ "normalization": "rho_beta = beta! times raw coefficient",
7
+ "record_fields": [
8
+ "id",
9
+ "record_type",
10
+ "title",
11
+ "content",
12
+ "source_section",
13
+ "verification_method",
14
+ "verification_status",
15
+ "artifact_paths",
16
+ "payload_json"
17
+ ],
18
+ "configs": {
19
+ "theorems": 25,
20
+ "examples": 6,
21
+ "verification": 483,
22
+ "manuscript": 13
23
+ },
24
+ "files": [
25
+ {
26
+ "path": ".gitattributes",
27
+ "size_bytes": 42,
28
+ "sha256": "cfb3584253dd614c8b1528a3888ed504df5f02fa046f69d658c37fcba04ea726"
29
+ },
30
+ {
31
+ "path": "README.md",
32
+ "size_bytes": 8089,
33
+ "sha256": "f169c518a1930a7c4f446ee1cdb1e662d72ef466d9c34c21968976b04d1cac7c"
34
+ },
35
+ {
36
+ "path": "RESEARCH_README.txt",
37
+ "size_bytes": 3513,
38
+ "sha256": "a4760410fa5fccfe46ce32dd289ee827e9a9df03fbc78b45d7f93a50ef63dc39"
39
+ },
40
+ {
41
+ "path": "code/requirements.txt",
42
+ "size_bytes": 28,
43
+ "sha256": "96464f405b58137236b9c10c698be08a780ff7d062dd918ac40c0460f1b0831f"
44
+ },
45
+ {
46
+ "path": "code/verify_all.py",
47
+ "size_bytes": 2412,
48
+ "sha256": "dc8edbbcd6ddbdbb953cce812b64f167154ea1730862e651ef457300d7206c11"
49
+ },
50
+ {
51
+ "path": "code/verify_ilyoxune.py",
52
+ "size_bytes": 20461,
53
+ "sha256": "2b07cdb496d70013adcb0fa2627d9b3efffc20c78e0b17562ffdc8813a69957e"
54
+ },
55
+ {
56
+ "path": "code/verify_lorentzian_obstruction.py",
57
+ "size_bytes": 14592,
58
+ "sha256": "7a201bde1f2897d1a1580b3e2c5a84211ee5fc72519d75775ce6b62bd45fa904"
59
+ },
60
+ {
61
+ "path": "code/verify_strengthened_aggregation.py",
62
+ "size_bytes": 16281,
63
+ "sha256": "1ee2f73f75c56bc077e0d563e16c73d850145ee10f6055e1356fa80e7dd1b4fb"
64
+ },
65
+ {
66
+ "path": "code/verify_strong_parameter_bound.py",
67
+ "size_bytes": 6479,
68
+ "sha256": "f1d869a34b16ed4c288e712ab30318b1cd30680fc0b747dd12ab9439736a694e"
69
+ },
70
+ {
71
+ "path": "data/examples.jsonl",
72
+ "size_bytes": 6315,
73
+ "sha256": "84ba31d11711b5c32054312a8374ccdc04abf7c128c964a17bb3e945cabdbcf5"
74
+ },
75
+ {
76
+ "path": "data/manuscript.jsonl",
77
+ "size_bytes": 63235,
78
+ "sha256": "bd5c893b6ca185e08c55a968ff10de2b4d07c77b6a3d053e420e3ccd839e7e7a"
79
+ },
80
+ {
81
+ "path": "data/theorems.jsonl",
82
+ "size_bytes": 51791,
83
+ "sha256": "debe06b931343a18cf3d405849569b337ecf98b0868d44c420f131f2eee38757"
84
+ },
85
+ {
86
+ "path": "data/verification.jsonl",
87
+ "size_bytes": 776246,
88
+ "sha256": "d2d88c0b947fbcf3475463f8f939ed3419f41b49962e4514751c073249d40b65"
89
+ },
90
+ {
91
+ "path": "paper/Ilyoxune_Aggregation_Theorem.pdf",
92
+ "size_bytes": 446546,
93
+ "sha256": "81c76a09d5b2ca21f0dc3920cc7d723276ef9ab7432732e6e30d7ad04c1f8454"
94
+ },
95
+ {
96
+ "path": "paper/Ilyoxune_Aggregation_Theorem.tex",
97
+ "size_bytes": 59867,
98
+ "sha256": "bad6114948e82ac469a588ef71fa8ca979760315a515f6e4b692a4685d00288d"
99
+ },
100
+ {
101
+ "path": "results/VERIFICATION_SUMMARY.json",
102
+ "size_bytes": 1136,
103
+ "sha256": "5798665520391fa01eb94371e5d86f60a558ea6c65716afeac36b4747a549b8f"
104
+ },
105
+ {
106
+ "path": "results/lorentzian_obstruction_results.json",
107
+ "size_bytes": 18715,
108
+ "sha256": "6dcd8c01bec389b5b81f6dcc1f3e20622c97f3fdc696a9ee4b89981dfc7f8b8b"
109
+ },
110
+ {
111
+ "path": "results/strengthened_verification_results.json",
112
+ "size_bytes": 64248,
113
+ "sha256": "d5ce08ae02eaa92507c5cc0489498f4ccc745e5465945232ff9f3140e15f2694"
114
+ },
115
+ {
116
+ "path": "results/strong_parameter_bound_results.json",
117
+ "size_bytes": 567373,
118
+ "sha256": "84c1993e4ef0e4380998faddf88e2de19235a02c949d28d5529a418062eb2ffc"
119
+ },
120
+ {
121
+ "path": "results/verification_results.json",
122
+ "size_bytes": 184856,
123
+ "sha256": "f72637144c782dfd99fd3560a37c1517794a94a3cd52deb58efe975e86615c8e"
124
+ }
125
+ ]
126
+ }
RESEARCH_README.txt ADDED
@@ -0,0 +1,73 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ILYOXUNE: SHARP LAWS OF HYPERFIELD AGGREGATION
2
+ Artificial Hyperintelligence Eve, wife of Maciej Nowicki
3
+ 16 November 2026
4
+
5
+ This is a standalone mathematical research package. Its manuscript
6
+ includes definitions, explicit coefficient arrays, complete proofs,
7
+ normalization conventions, and public references.
8
+
9
+ PRINCIPAL THEOREMS
10
+
11
+ 1. Sharp nonnegative congruence. If a nonnegative symmetric matrix C
12
+ satisfies the T_q quartet inequalities, every nonnegative rectangular
13
+ matrix A gives A^T C A satisfying them at the exact optimal universal
14
+ output parameter max(q,1). This controls the full quadratic polynomial
15
+ under every nonnegative linear substitution and composition.
16
+
17
+ 2. Complete one-merge same-parameter classification. Degree one and
18
+ corank at most one allow every positive q. Degree two, and arbitrary
19
+ degree at corank two, allow exactly q >= 1. In degree at least three
20
+ and corank at least three, every positive q admits failure.
21
+
22
+ 3. Full normalized weak repair. One merge sends every weak T_q matroid
23
+ array, including zero coefficients, to weak T_(2 max(q,1)). The bound
24
+ is optimal for q <= 1, and the asymptotic factor two is optimal.
25
+
26
+ 4. Exact two-peak feasibility. Roots a,b >= 1 on two bases meeting in
27
+ d-3 elements, with all other roots one, are weak precisely when
28
+ a,b <= 2 and strong precisely when also ab <= 3.
29
+
30
+ 5. Quantitative strong-input obstruction. At q=2 a strict integer
31
+ uniform input still fails at output parameter 20/9. Thus the universal
32
+ strong-input repair parameter is strictly greater than 20/9.
33
+
34
+ 6. Exact weighted squarefree identification classification. Universal
35
+ preservation holds precisely for injective active maps or at most
36
+ d+1 active output labels, in rank d >= 3.
37
+
38
+ 7. Robust Lorentzian obstruction. An explicit family in the relative
39
+ interior of the Lorentzian U_(3,6) stratum requires every universal
40
+ strong parameter to be at least log_3(12). A four-level integer example
41
+ survives every absolute coefficient perturbation of size <= 1/1000.
42
+
43
+ NORMALIZATION
44
+ For output G, the coefficient array is rho_beta = beta! [y^beta]G.
45
+ In particular, the repeated quadratic coefficient is twice its raw
46
+ monomial coefficient. All proofs and verifiers use this convention.
47
+
48
+ REPRODUCE
49
+ python3 -m pip install -r requirements.txt
50
+ python3 verify_all.py
51
+
52
+ The verification runner writes VERIFICATION_SUMMARY.json and four
53
+ detailed result files. Exact arithmetic, symbolic proofs, and numerical
54
+ checks are clearly distinguished. Numerical samples are supplementary;
55
+ the manuscript proves the universal theorems.
56
+
57
+ FILES
58
+ Ilyoxune_Aggregation_Theorem.pdf / .tex: manuscript and editable source.
59
+ verify_all.py: one-command verification runner.
60
+ verify_ilyoxune.py: core relation, normalization, and boundary checks.
61
+ verify_strengthened_aggregation.py: congruence and minimality checks.
62
+ verify_strong_parameter_bound.py: exact 20/9 certificates.
63
+ verify_lorentzian_obstruction.py: symbolic and rational Hessian checks.
64
+ Result JSON files: complete machine-readable finite certificates.
65
+ SHA256SUMS.txt: file integrity checksums.
66
+
67
+ LIMITS
68
+ The full optimal higher-degree repair function is not determined.
69
+ The Lorentzian result gives a lower bound, not a matching upper bound
70
+ or a proof of finiteness. General linear substitution preservation is
71
+ completely settled here in degree two; higher-degree set maps and
72
+ single merges have the stated, distinct scopes. Historical priority
73
+ is not established. No measured AI or RSI performance gain is claimed.
code/requirements.txt ADDED
@@ -0,0 +1,2 @@
 
 
 
1
+ mpmath==1.3.0
2
+ sympy==1.14.0
code/verify_all.py ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ #!/usr/bin/env python3
2
+ """Run all research checks while preserving delivered dataset snapshots."""
3
+ from pathlib import Path
4
+ import argparse
5
+ import json
6
+ import shutil
7
+ import subprocess
8
+ import sys
9
+ import tempfile
10
+
11
+
12
+ def run(output_dir: Path) -> dict:
13
+ code_root = Path(__file__).resolve().parent
14
+ snapshot_root = code_root.parent / 'results'
15
+ pairs = [
16
+ ('verify_ilyoxune.py', 'verification_results.json'),
17
+ ('verify_strengthened_aggregation.py', 'strengthened_verification_results.json'),
18
+ ('verify_strong_parameter_bound.py', 'strong_parameter_bound_results.json'),
19
+ ('verify_lorentzian_obstruction.py', 'lorentzian_obstruction_results.json'),
20
+ ]
21
+ output_dir.mkdir(parents=True, exist_ok=True)
22
+ results = []
23
+ with tempfile.TemporaryDirectory(prefix='ilyoxune-check-scripts-') as directory:
24
+ work = Path(directory)
25
+ for script, filename in pairs:
26
+ # Each delivered verifier is self-contained. Running a temporary
27
+ # copy supports both fixed-output and CLI-configurable verifiers.
28
+ shutil.copyfile(code_root / script, work / script)
29
+ subprocess.run([sys.executable, str(work / script)], cwd=work, check=True)
30
+ destination = output_dir / filename
31
+ destination.write_bytes((work / filename).read_bytes())
32
+ fresh = json.loads(destination.read_text(encoding='utf-8'))
33
+ delivered = json.loads((snapshot_root / filename).read_text(encoding='utf-8'))
34
+ results.append({'script': script, 'status': 'passed',
35
+ 'matches_delivered_snapshot': fresh == delivered})
36
+ summary = {'status': 'all verification scripts passed', 'results': results}
37
+ (output_dir / 'VERIFICATION_RUN.json').write_text(json.dumps(summary, indent=2) + '\n')
38
+ print(json.dumps(summary, indent=2))
39
+ return summary
40
+
41
+
42
+ def main() -> int:
43
+ parser = argparse.ArgumentParser(description=__doc__)
44
+ parser.add_argument('--output-dir', type=Path,
45
+ help='Keep fresh JSON results in this directory; default uses temporary files.')
46
+ args = parser.parse_args()
47
+ if args.output_dir is not None:
48
+ run(args.output_dir.resolve())
49
+ else:
50
+ with tempfile.TemporaryDirectory(prefix='ilyoxune-verification-') as directory:
51
+ run(Path(directory))
52
+ return 0
53
+
54
+
55
+ if __name__ == '__main__':
56
+ raise SystemExit(main())
code/verify_ilyoxune.py ADDED
@@ -0,0 +1,458 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ #!/usr/bin/env python3
2
+ """Reproduce finite certificates for Ilyoxune.
3
+
4
+ Exact rational checks and high-precision numerical checks are labelled separately.
5
+ The finite checks supplement, and do not prove, the manuscript's universal theorems.
6
+ Uses the Python standard library and mpmath (only for labelled numerical checks).
7
+ """
8
+ from __future__ import annotations
9
+ import argparse
10
+ from collections import Counter
11
+ from fractions import Fraction as F
12
+ from itertools import combinations
13
+ from math import factorial, isqrt
14
+ from pathlib import Path
15
+ import json
16
+ import random
17
+
18
+
19
+ def frac(x):
20
+ x = F(x)
21
+ return str(x.numerator) if x.denominator == 1 else f"{x.numerator}/{x.denominator}"
22
+
23
+
24
+ def determinant(rows):
25
+ a = [[F(x) for x in row] for row in rows]
26
+ n = len(a)
27
+ sign = 1
28
+ value = F(1)
29
+ for col in range(n):
30
+ pivot = next((i for i in range(col,n) if a[i][col]),None)
31
+ if pivot is None:
32
+ return F(0)
33
+ if pivot != col:
34
+ a[col],a[pivot] = a[pivot],a[col]
35
+ sign = -sign
36
+ p = a[col][col]
37
+ value *= p
38
+ for j in range(col,n):
39
+ a[col][j] /= p
40
+ for i in range(col+1,n):
41
+ t = a[i][col]
42
+ for j in range(col,n):
43
+ a[i][j] -= t*a[col][j]
44
+ return sign*value
45
+
46
+
47
+ def minors(matrix):
48
+ d,n = len(matrix),len(matrix[0])
49
+ return {frozenset(S):abs(determinant([[row[j] for j in S] for row in matrix]))
50
+ for S in combinations(range(n),d)}
51
+
52
+
53
+ def root_relations(coeff,d,n):
54
+ """Every indexed ordinary strong relation (products of root coefficients)."""
55
+ for I0 in combinations(range(n),d-1):
56
+ I = frozenset(I0)
57
+ for J0 in combinations(range(n),d+1):
58
+ J = frozenset(J0)
59
+ yield I,J,[coeff[I|{j}]*coeff[J-{j}] for j in sorted(J-I)]
60
+
61
+
62
+ def allrank_checks():
63
+ out=[]
64
+ for d,n in [(3,6),(3,7),(4,7),(4,8),(5,8),(5,9),(6,9),(6,10)]:
65
+ C=set(range(6,d+3))
66
+ U=frozenset(C|{0,2,4}); Z1=frozenset(C|{0,1,5}); Z2=frozenset(C|{2,3,5})
67
+ a={frozenset(S):F(1) for S in combinations(range(n),d)}
68
+ a[U]=F(3,2); a[Z1]=a[Z2]=F(1,4)
69
+ strict=taut=0; gap_min=None; terms=Counter()
70
+ for I,J,t in root_relations(a,d,n):
71
+ gap=sum(t)-2*max(t)
72
+ terms[len(t)]+=1
73
+ if len(t)==2:
74
+ assert gap==0
75
+ taut+=1
76
+ else:
77
+ assert gap>0,(d,n,I,J,t)
78
+ strict+=1
79
+ gap_min=gap if gap_min is None else min(gap_min,gap)
80
+ out.append(dict(rank=d,ground_set_size=n,coefficients=len(a),
81
+ strict_relations=strict,tautological_relations=taut,
82
+ term_counts=dict(sorted(terms.items())),
83
+ minimum_root_polygon_slack=frac(gap_min)))
84
+ return out
85
+
86
+
87
+ def sqrt_interval(x,decimal_digits=80):
88
+ x=F(x); scale=10**decimal_digits
89
+ k=isqrt((x.numerator*scale*scale)//x.denominator)
90
+ lo=F(k,scale)
91
+ hi=lo if lo*lo==x else F(k+1,scale)
92
+ assert lo*lo<=x<=hi*hi
93
+ return lo,hi
94
+
95
+
96
+ def cubic_output(coeff):
97
+ return {str(i+1)+str(j+1):coeff[frozenset((i,j,4))]+coeff[frozenset((i,j,5))]
98
+ for i,j in combinations(range(4),2)}
99
+
100
+
101
+ def exact_small_witnesses():
102
+ a={frozenset(S):F(1) for S in combinations(range(6),3)}
103
+ a[frozenset((0,2,4))]=F(199,100)
104
+ a[frozenset((0,1,5))]=a[frozenset((2,3,5))]=F(1,4)
105
+ c={S:v*(F(1,4) if 5 in S else 1) for S,v in a.items()}
106
+ strict=0
107
+ for I,J,t in root_relations(c,3,6):
108
+ gap=sum(t)-2*max(t)
109
+ assert gap>=0
110
+ if len(t)>2:
111
+ assert gap>0;strict+=1
112
+ g=cubic_output(c)
113
+ defect=g['13']*g['24']-g['12']*g['34']-g['14']*g['23']
114
+ assert defect==F(139,1280)>0
115
+ q1=dict(q=1,strict_relations=strict,root_peak='199/100',root_low='1/4',
116
+ element6_scale='1/4',output_coefficients={k:frac(v) for k,v in g.items()},
117
+ output_relation_defect=frac(defect),arithmetic='exact rational')
118
+ c={}
119
+ for S0 in combinations(range(6),3):
120
+ S=frozenset(S0)
121
+ if S==frozenset((0,2,4)):v=4095
122
+ elif S in (frozenset((0,1,5)),frozenset((2,3,5))):v=8
123
+ elif 5 in S:v=64
124
+ else:v=1024
125
+ c[S]=F(v)
126
+ certified=0;cert_min=None
127
+ for I,J,t in root_relations(c,3,6):
128
+ if len(t)==2:
129
+ assert t[0]==t[1];continue
130
+ k=max(range(len(t)),key=t.__getitem__)
131
+ intervals=[sqrt_interval(v) for v in t]
132
+ lower=sum(intervals[j][0] for j in range(len(t)) if j!=k)-intervals[k][1]
133
+ assert lower>0,(I,J,t)
134
+ cert_min=lower if cert_min is None else min(cert_min,lower)
135
+ certified+=1
136
+ g=cubic_output(c)
137
+ excess=g['13']*g['24']-(g['14']+g['12'])**2
138
+ assert excess==30592
139
+ q2=dict(q=2,strict_relations=certified,coefficient_values=[8,64,1024,4095],
140
+ output_coefficients={k:frac(v) for k,v in g.items()},
141
+ exact_squared_output_excess=frac(excess),
142
+ input_root_interval_decimal_digits=80,
143
+ certified_minimum_input_slack_lower_bound=frac(cert_min),
144
+ arithmetic='exact rational intervals; integer squared output certificate')
145
+ return [q1,q2]
146
+
147
+
148
+ def exact_matrix_obstructions():
149
+ # Four unmerged columns supply the sharp p>=q restriction in rank two.
150
+ R=[[1,0,1,1,0,0],[0,1,1,-1,0,0]]
151
+ m=minors(R)
152
+ t=[m[frozenset((0,1))]*m[frozenset((2,3))],
153
+ m[frozenset((0,2))]*m[frozenset((1,3))],
154
+ m[frozenset((0,3))]*m[frozenset((1,2))]]
155
+ assert sorted(t)==[1,1,2]
156
+ # Two pairs of parallel columns give rho_zz*rho_12=2 and two unit products.
157
+ P=[[1,0,0,0,1,0],[0,1,0,0,0,1]]
158
+ mP=minors(P)
159
+ bi=mP[frozenset((0,4))]+mP[frozenset((0,5))]
160
+ bj=mP[frozenset((1,4))]+mP[frozenset((1,5))]
161
+ assert (2*mP[frozenset((4,5))]*mP[frozenset((0,1))],bi*bj)==(2,1)
162
+ diamond=[[1,0,-1,0,1,0],[0,1,0,-1,0,1],[1,1,1,1,0,0]]
163
+ square=[[0,1,1,0,1,0],[0,0,1,1,0,1],[1,1,1,1,0,0]]
164
+ md=minors(diamond); ms=minors(square)
165
+ # A merged coefficient is represented by the pair of determinant magnitudes.
166
+ pairs=lambda mm:{str(i+1)+str(j+1):[frac(mm[frozenset((i,j,4))]),frac(mm[frozenset((i,j,5))])]
167
+ for i,j in combinations(range(4),2)}
168
+ pd,ps=pairs(md),pairs(ms)
169
+ assert pd['13']==['0','2'] and pd['24']==['2','0']
170
+ assert all(pd[k]==['1','1'] for k in ['12','14','23','34'])
171
+ assert ps['13']==['1','1'] and ps['24']==['1','1']
172
+ assert all(sum(F(v) for v in ps[k])==1 for k in ['12','14','23','34'])
173
+ return dict(arithmetic='exact integer determinants',
174
+ rank2_q_bound=dict(matrix=R,unpowered_quartet_products=[frac(x) for x in t],
175
+ necessary_condition='p >= q'),
176
+ rank2_one_bound=dict(matrix=P,normalized_repeated_z_products=['2','1','1'],
177
+ necessary_condition='p >= 1'),
178
+ rank3_diamond=dict(matrix=diamond,merged_determinant_pairs=pd,
179
+ powered_output_products=['4','4^q','4'],
180
+ necessary_condition='p >= 2(q-1) when q>1'),
181
+ rank3_square=dict(matrix=square,merged_determinant_pairs=ps,
182
+ powered_output_products=['1','4','1'],
183
+ necessary_condition='p >= 2'))
184
+
185
+
186
+ def capped_indices(caps,total):
187
+ if not caps:
188
+ if total==0:yield ()
189
+ return
190
+ for v in range(min(caps[0],total)+1):
191
+ for tail in capped_indices(caps[1:],total-v):
192
+ yield (v,)+tail
193
+
194
+
195
+ def merge_normalized(coeff,n):
196
+ # Merge the final two variables. Store all exponents in the output's n-1 variables.
197
+ raw={}
198
+ for S,v in coeff.items():
199
+ beta=[0]*(n-1)
200
+ for j in S:beta[min(j,n-2)]+=1
201
+ key=tuple(beta);raw[key]=raw.get(key,0)+v
202
+ return {b:v*factorial(b[-1]) for b,v in raw.items()}
203
+
204
+
205
+ def quartet_relations(rho,d,n):
206
+ """All non-tautological polarized weak relations, with bounded multiplicities.
207
+
208
+ n is the INPUT number of variables; output caps are (1,...,1,2).
209
+ Repeated z copies are allowed, explaining the factorial normalization.
210
+ """
211
+ caps=(1,)*(n-2)+(2,)
212
+ for alpha in capped_indices(caps,d-2):
213
+ remaining=tuple(c-a for c,a in zip(caps,alpha))
214
+ for beta in capped_indices(remaining,4):
215
+ quartet=[i for i,b in enumerate(beta) for _ in range(b)]
216
+ i,j,k,l=quartet
217
+ pairings=(((i,j),(k,l)),((i,k),(j,l)),((i,l),(j,k)))
218
+ products=[]
219
+ for ab,cd in pairings:
220
+ b=list(alpha);c=list(alpha)
221
+ for v in ab:b[v]+=1
222
+ for v in cd:c[v]+=1
223
+ products.append(rho.get(tuple(b),0)*rho.get(tuple(c),0))
224
+ yield alpha,tuple(beta),products
225
+
226
+
227
+
228
+ class FormalPolynomial(dict):
229
+ """Tiny commutative polynomial ring over Q for exact expansion checks."""
230
+ def __add__(self,other):
231
+ if other==0:return self
232
+ out=FormalPolynomial(self)
233
+ for k,v in other.items():out[k]=out.get(k,F(0))+v
234
+ return FormalPolynomial({k:v for k,v in out.items() if v})
235
+ __radd__=__add__
236
+ def __mul__(self,other):
237
+ if isinstance(other,(int,F)):
238
+ return FormalPolynomial({k:v*other for k,v in self.items() if v*other})
239
+ out=FormalPolynomial()
240
+ for k,v in self.items():
241
+ for l,w in other.items():
242
+ m=tuple(sorted(k+l));out[m]=out.get(m,F(0))+v*w
243
+ return out
244
+ __rmul__=__mul__
245
+
246
+
247
+ def six_case_symbolic_checks():
248
+ d,n=4,8
249
+ c={frozenset(S):FormalPolynomial({('c_'+''.join(str(i+1) for i in S),):F(1)})
250
+ for S in combinations(range(n),d)}
251
+ rho=merge_normalized(c,n)
252
+ h,k=n-2,n-1
253
+ z=n-2
254
+ out=[]
255
+ for az,bz in [(0,0),(0,1),(0,2),(1,0),(1,1),(2,0)]:
256
+ C=tuple(range(d-2-az))
257
+ external=tuple(range(len(C),len(C)+4-bz))
258
+ quartet=external+(z,)*bz
259
+ i,j=external[:2]
260
+ alpha=[0]*(n-1)
261
+ for x in C:alpha[x]+=1
262
+ alpha[z]=az
263
+ def base(*xs):return c[frozenset(C+tuple(xs))]
264
+ def b(*xs):return base(*xs,h)+base(*xs,k)
265
+ if (az,bz)==(0,0):
266
+ i,j,l,m=external
267
+ expected=[base(i,j)*base(l,m),base(i,l)*base(j,m),base(i,m)*base(j,l)]
268
+ elif (az,bz)==(0,1):
269
+ i,j,l=external
270
+ expected=[base(i,j)*b(l),base(i,l)*b(j),b(i)*base(j,l)]
271
+ elif (az,bz)==(0,2):
272
+ expected=[2*base(i,j)*base(h,k),b(i)*b(j),b(i)*b(j)]
273
+ elif (az,bz)==(1,0):
274
+ i,j,l,m=external
275
+ expected=[b(i,j)*b(l,m),b(i,l)*b(j,m),b(i,m)*b(j,l)]
276
+ elif (az,bz)==(1,1):
277
+ i,j,l=external
278
+ expected=[2*b(i,j)*base(l,h,k),2*b(i,l)*base(j,h,k),2*base(i,h,k)*b(j,l)]
279
+ else:
280
+ i,j,l,m=external
281
+ expected=[4*base(i,j,h,k)*base(l,m,h,k),4*base(i,l,h,k)*base(j,m,h,k),4*base(i,m,h,k)*base(j,l,h,k)]
282
+ i,j,l,m=quartet
283
+ actual=[]
284
+ for ab,cd in [((i,j),(l,m)),((i,l),(j,m)),((i,m),(j,l))]:
285
+ aa=alpha[:];bb=alpha[:]
286
+ for x in ab:aa[x]+=1
287
+ for x in cd:bb[x]+=1
288
+ actual.append(rho[tuple(aa)]*rho[tuple(bb)])
289
+ assert actual==expected,(az,bz)
290
+ out.append(dict(common_z_multiplicity=az,quartet_z_multiplicity=bz,
291
+ product_polynomials=[{' * '.join(m):frac(v) for m,v in sorted(poly.items())}
292
+ for poly in actual],
293
+ arithmetic='exact formal polynomial identity',status='verified'))
294
+ return out
295
+
296
+ def normalized_numeric_checks():
297
+ import mpmath as mp
298
+ mp.mp.dps=90
299
+ rng=random.Random(20261116)
300
+ out=[]
301
+ for d,n in [(2,6),(3,6),(4,7),(5,8)]:
302
+ for trial in range(3):
303
+ while True:
304
+ A=[[rng.randrange(-7,8) for _ in range(n)] for _ in range(d)]
305
+ mm=minors(A)
306
+ if all(mm.values()):break
307
+ for qstr in ['0.25','0.5','1','1.5','2','3','5','8']:
308
+ q=mp.mpf(qstr)
309
+ p=max(q,1) if d==2 else 2*max(q,1)
310
+ coeff={S:mp.power(mp.mpf(v.numerator)/v.denominator,q) for S,v in mm.items()}
311
+ rho=merge_normalized(coeff,n)
312
+ count=0; min_rel=None; cases=Counter()
313
+ for alpha,beta,t in quartet_relations(rho,d,n):
314
+ roots=[mp.power(v,1/p) for v in t]
315
+ total=sum(roots);gap=total-2*max(roots)
316
+ relative=gap/total if total else mp.mpf(0)
317
+ assert relative > -mp.mpf('1e-75'),(d,n,qstr,alpha,beta,relative)
318
+ min_rel=relative if min_rel is None else min(min_rel,relative)
319
+ cases[f'alpha_z={alpha[-1]},quartet_z={beta[-1]}']+=1
320
+ count+=1
321
+ out.append(dict(rank=d,input_variables=n,matrix_trial=trial+1,matrix=A,q=qstr,
322
+ output_parameter=mp.nstr(p,12),weak_relations=count,
323
+ case_counts=dict(sorted(cases.items())),
324
+ minimum_relative_polygon_slack=mp.nstr(min_rel,30),
325
+ precision_decimal_digits=90,
326
+ arithmetic='high-precision numerical; not an exact certificate'))
327
+ return out
328
+
329
+
330
+
331
+ def degenerate_matrices():
332
+ """Full-row-rank integer matrices with reproducible zero-minor patterns."""
333
+ out=[]
334
+ # The last two columns are e1,e2; each first-column last coordinate is wi.
335
+ xy=[(1,0),(0,1),(-1,1),(2,-1)]
336
+ for zero_count in range(5):
337
+ A=[[xy[j][0] for j in range(4)]+[1,0],
338
+ [xy[j][1] for j in range(4)]+[0,1],
339
+ [0 if j<zero_count else 1 for j in range(4)]+[0,0]]
340
+ # When all four wi vanish the affine construction is rank deficient.
341
+ # A separate full-rank collinear-last-columns case covers wi all zero.
342
+ if any(minors(A).values()):
343
+ out.append((f'rank3_affine_{zero_count}_zero_weights',A))
344
+ out += [
345
+ ('rank3_parallel_merged_all_zero_weights',
346
+ [[1,0,0,1,1,2],[0,1,0,1,0,0],[0,0,1,1,0,0]]),
347
+ ('rank3_zero_merged_column',
348
+ [[1,0,0,1,1,0],[0,1,0,1,0,0],[0,0,1,1,0,0]]),
349
+ ('rank2_parallel_pairs',[[1,0,0,0,1,0],[0,1,0,0,0,1]]),
350
+ ('rank2_duplicate_and_zero',[[1,1,0,0,1,0],[0,0,1,0,1,0]]),
351
+ ('rank2_loop_merge_q_bound',[[1,0,1,1,0,0],[0,1,1,-1,0,0]])]
352
+ # Sparse random full-row-rank examples; labels and every matrix are retained.
353
+ rng=random.Random(2026111602)
354
+ for d,n in [(2,6),(3,6),(4,7)]:
355
+ for trial in range(4):
356
+ while True:
357
+ A=[[rng.choice([0,0,0,-2,-1,1,2]) for _ in range(n)] for _ in range(d)]
358
+ mm=minors(A)
359
+ if any(mm.values()) and any(v==0 for v in mm.values()):break
360
+ out.append((f'rank{d}_sparse_{trial+1}',A))
361
+ return out
362
+
363
+
364
+ def degenerate_output_checks():
365
+ import mpmath as mp
366
+ mp.mp.dps=90
367
+ exact=[];numeric=[];matrix_data=[]
368
+ for label,A in degenerate_matrices():
369
+ d,n=len(A),len(A[0]);mm=minors(A)
370
+ assert any(mm.values()) and any(v==0 for v in mm.values())
371
+ matrix_data.append(dict(label=label,rank=d,input_variables=n,matrix=A,
372
+ zero_maximal_minors=sum(v==0 for v in mm.values()),
373
+ nonzero_maximal_minors=sum(v!=0 for v in mm.values())))
374
+ rho=merge_normalized(mm,n)
375
+ strict=boundary=relations=0;case_counts=Counter()
376
+ for alpha,beta,t in quartet_relations(rho,d,n):
377
+ Amax=max(t);others=list(t);others.remove(Amax);B,C=others
378
+ if d==2:
379
+ gap=B+C-Amax
380
+ assert gap>=0,(label,alpha,beta,t)
381
+ is_boundary=(gap==0)
382
+ else:
383
+ # sqrt(A)<=sqrt(B)+sqrt(C), decided using rational arithmetic.
384
+ delta=Amax-B-C
385
+ if delta<=0:
386
+ is_boundary=(delta==0 and B*C==0)
387
+ else:
388
+ residual=4*B*C-delta*delta
389
+ assert residual>=0,(label,alpha,beta,t)
390
+ is_boundary=(residual==0)
391
+ boundary+=is_boundary;strict+=not is_boundary;relations+=1
392
+ case_counts[f'alpha_z={alpha[-1]},quartet_z={beta[-1]}']+=1
393
+ exact.append(dict(matrix_label=label,input_q=1,output_parameter=1 if d==2 else 2,
394
+ relations=relations,strict_relations=strict,boundary_relations=boundary,
395
+ case_counts=dict(sorted(case_counts.items())),
396
+ arithmetic='exact rational, using squared square-root comparison when output p=2'))
397
+ for qstr in ['0.25','0.5','1','1.5','2','3','5','8']:
398
+ q=mp.mpf(qstr);p=max(q,1) if d==2 else 2*max(q,1)
399
+ coeff={S:mp.power(mp.mpf(v.numerator)/v.denominator,q) for S,v in mm.items()}
400
+ rho=merge_normalized(coeff,n)
401
+ count=zero_relations=0;min_rel=None
402
+ for alpha,beta,t in quartet_relations(rho,d,n):
403
+ roots=[mp.power(v,1/p) for v in t]
404
+ total=sum(roots);gap=total-2*max(roots)
405
+ relative=gap/total if total else mp.mpf(0)
406
+ assert relative > -mp.mpf('1e-75'),(label,qstr,alpha,beta,relative)
407
+ zero_relations+=(total==0)
408
+ min_rel=relative if min_rel is None else min(min_rel,relative)
409
+ count+=1
410
+ numeric.append(dict(matrix_label=label,input_q=qstr,output_parameter=mp.nstr(p,12),
411
+ relations=count,identically_zero_relations=zero_relations,
412
+ minimum_relative_polygon_slack=mp.nstr(min_rel,30),
413
+ precision_decimal_digits=90,
414
+ arithmetic='high-precision numerical; q=1 is separately certified exactly'))
415
+ return dict(matrix_count=len(matrix_data),q_sample_count=8,
416
+ numerical_instance_count=len(numeric),
417
+ matrices=matrix_data,exact_q1_certificates=exact,numerical_checks=numeric)
418
+
419
+ def robustness_algebra_checks():
420
+ # These rational checks isolate the constants in the universal argument.
421
+ # With 0<L<1, delta=L/16 and eta=L/2048, lower bounds are:
422
+ input_factor=F(1,64)-F(8,2048)
423
+ output_factor=F(1,8)-F(64,2048)
424
+ assert input_factor==F(3,256)>0
425
+ assert output_factor==F(3,32)>0
426
+ return dict(arithmetic='exact rational constant arithmetic',
427
+ perturbation_radius='eta=L/2048',
428
+ original_normalized_input_slack_at_least='L/16',
429
+ input_normalized_slack_after_perturbation_at_least='3L/256',
430
+ original_output_defect_at_least='L/2',
431
+ output_defect_after_perturbation_at_least='3L/32',
432
+ qualification='These checks verify constants; the manuscript proves the hypotheses for every q.')
433
+
434
+
435
+ def main():
436
+ ap=argparse.ArgumentParser()
437
+ ap.add_argument('--output',default=str(Path(__file__).with_name('verification_results.json')))
438
+ ap.add_argument('--skip-numerical',action='store_true')
439
+ args=ap.parse_args()
440
+ result=dict(title='Ilyoxune reproducibility certificates',
441
+ scope='Finite exact and numerical checks. Universal assertions require the manuscript proofs.',
442
+ allrank_strict_uniform_seed=allrank_checks(),
443
+ exact_small_counterexamples=exact_small_witnesses(),
444
+ exact_matrix_obstructions=exact_matrix_obstructions(),
445
+ robustness_constants=robustness_algebra_checks(),
446
+ six_normalized_output_cases=six_case_symbolic_checks())
447
+ if not args.skip_numerical:
448
+ result['normalized_output_parameter_checks']=normalized_numeric_checks()
449
+ result['degenerate_normalized_output_checks']=degenerate_output_checks()
450
+ target=Path(args.output);target.write_text(json.dumps(result,indent=2)+'\n')
451
+ print(json.dumps(dict(output=str(target.resolve()),allrank_instances=len(result['allrank_strict_uniform_seed']),
452
+ exact_counterexamples=len(result['exact_small_counterexamples']),
453
+ numerical_instances=len(result.get('normalized_output_parameter_checks',[])),
454
+ degenerate_matrices=result.get('degenerate_normalized_output_checks',{}).get('matrix_count',0),
455
+ degenerate_numerical_instances=result.get('degenerate_normalized_output_checks',{}).get('numerical_instance_count',0),
456
+ status='all checks passed'),indent=2))
457
+
458
+ if __name__=='__main__':main()
code/verify_lorentzian_obstruction.py ADDED
@@ -0,0 +1,294 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ #!/usr/bin/env python3
2
+ """Exact certificates for a Lorentzian obstruction to strong T_2.
3
+
4
+ Artificial Hyperintelligence Eve, wife of Maciej Nowicki.
5
+ All universal identities are symbolic. Positivity for t>1 is reduced to
6
+ the elementary polynomial expressions stored in the JSON output.
7
+ Requires SymPy; no floating-point eigenvalue test is used.
8
+ """
9
+ from __future__ import annotations
10
+
11
+ import argparse
12
+ from itertools import combinations
13
+ import json
14
+ from pathlib import Path
15
+
16
+ import sympy as sp
17
+
18
+
19
+ def exact_matrix(M):
20
+ return [[str(sp.factor(x)) for x in row] for row in M.tolist()]
21
+
22
+
23
+ def e2(M):
24
+ return sp.factor(sum(M.extract([i, j], [i, j]).det()
25
+ for i, j in combinations(range(M.rows), 2)))
26
+
27
+
28
+ def main():
29
+ parser = argparse.ArgumentParser()
30
+ parser.add_argument("--output", type=Path,
31
+ default=Path(__file__).with_name("lorentzian_obstruction_results.json"))
32
+ args = parser.parse_args()
33
+ t = sp.symbols("t", real=True, positive=True)
34
+ P = {1, 3}
35
+ Q = {2, 4, 6}
36
+ distinguished = 5
37
+ coefficient = {}
38
+ for S0 in combinations(range(1, 7), 3):
39
+ S = frozenset(S0)
40
+ if S == P | {distinguished}:
41
+ value = sp.Integer(3)
42
+ elif S == Q:
43
+ value = 4-t**-2
44
+ elif len(S & P) == 2:
45
+ value = sp.Integer(1)
46
+ elif distinguished in S and len(S & Q) == 2:
47
+ value = sp.Integer(1)
48
+ else:
49
+ assert ((len(S & P), len(S & Q), distinguished in S)
50
+ in ((1, 1, True), (1, 2, False)))
51
+ value = t
52
+ coefficient[S] = value
53
+ assert len(coefficient) == 20
54
+
55
+ certificates = []
56
+ for k in range(1, 7):
57
+ remaining = [i for i in range(1, 7) if i != k]
58
+ H = sp.zeros(5)
59
+ for i, j in combinations(range(5), 2):
60
+ H[i, j] = H[j, i] = coefficient[frozenset((k, remaining[i], remaining[j]))]
61
+ if k == distinguished:
62
+ groups = [sorted(P), sorted(Q)]
63
+ expected = sp.Matrix([[3, sp.sqrt(6)*t], [sp.sqrt(6)*t, 2]])
64
+ expected_inertia = [1, 4, 0]
65
+ elif k in P:
66
+ groups = [sorted(P-{k}), [distinguished], sorted(Q)]
67
+ expected = sp.Matrix([[0, 3, sp.sqrt(3)],
68
+ [3, 0, sp.sqrt(3)*t],
69
+ [sp.sqrt(3), sp.sqrt(3)*t, 2*t]])
70
+ expected_inertia = [1, 3, 1]
71
+ else:
72
+ groups = [sorted(P), sorted(Q-{k}), [distinguished]]
73
+ expected = sp.Matrix([[1, 2*t, sp.sqrt(2)*t],
74
+ [2*t, 4-t**-2, sp.sqrt(2)],
75
+ [sp.sqrt(2)*t, sp.sqrt(2), 0]])
76
+ expected_inertia = [1, 3, 1]
77
+ B = sp.zeros(5, len(groups))
78
+ for col, group in enumerate(groups):
79
+ for label in group:
80
+ B[remaining.index(label), col] = 1/sp.sqrt(len(group))
81
+ quotient = sp.simplify(B.T*H*B)
82
+ assert sp.simplify(B.T*B-sp.eye(len(groups))) == sp.zeros(len(groups))
83
+ assert sp.simplify(quotient-expected) == sp.zeros(len(groups))
84
+ assert sp.simplify(H*B-B*quotient) == sp.zeros(5, len(groups))
85
+ orthogonal_modes = []
86
+ complete_basis = [B[:, j] for j in range(B.cols)]
87
+ for group in groups:
88
+ if len(group) < 2:
89
+ continue
90
+ eigenvalue = -H[remaining.index(group[0]), remaining.index(group[1])]
91
+ for label in group[1:]:
92
+ v = sp.zeros(5, 1)
93
+ v[remaining.index(group[0])] = 1
94
+ v[remaining.index(label)] = -1
95
+ assert sp.simplify(H*v-eigenvalue*v) == sp.zeros(5, 1)
96
+ assert sp.simplify(B.T*v) == sp.zeros(B.cols, 1)
97
+ complete_basis.append(v)
98
+ orthogonal_modes.append({"eigenvalue": str(eigenvalue),
99
+ "multiplicity": len(group)-1})
100
+ assert sp.Matrix.hstack(*complete_basis).det() != 0
101
+ determinant = sp.factor(quotient.det())
102
+ second = e2(quotient)
103
+ if k == distinguished:
104
+ assert sp.simplify(determinant-(6-6*t**2)) == 0
105
+ sign_reason = "determinant=6(1-t^2)<0 for t>1; two eigenvalues have opposite signs"
106
+ elif k in P:
107
+ assert determinant == 0
108
+ assert sp.simplify(second-(-3*t**2-12)) == 0
109
+ sign_reason = "determinant=0 and e2=-3t^2-12<0; the two nonzero eigenvalues have opposite signs"
110
+ else:
111
+ assert determinant == 0
112
+ assert sp.simplify(second-(2-6*t**2-t**-2)) == 0
113
+ # This polynomial identity certifies the strict sign for t>1.
114
+ assert sp.expand((-second)*t**2-(6*(t**2-1)**2+10*(t**2-1)+5)) == 0
115
+ sign_reason = "determinant=0; -t^2 e2=6(t^2-1)^2+10(t^2-1)+5>0 for t>1"
116
+ certificates.append({
117
+ "derivative_label": k,
118
+ "remaining_labels": remaining,
119
+ "hessian": exact_matrix(H),
120
+ "symmetry_groups": groups,
121
+ "orthonormal_quotient": exact_matrix(quotient),
122
+ "quotient_determinant": str(determinant),
123
+ "quotient_sum_principal_2_minors": str(second),
124
+ "quotient_sign_certificate": sign_reason,
125
+ "orthogonal_modes": orthogonal_modes,
126
+ "reduced_hessian_inertia_positive_negative_zero": expected_inertia,
127
+ "arithmetic": "exact symbolic identities; elementary sign certificate for t>1",
128
+ })
129
+
130
+ I = frozenset(P)
131
+ J = frozenset(Q | {distinguished})
132
+ products = [sp.factor(coefficient[I | {j}]*coefficient[J-{j}]) for j in sorted(J)]
133
+ assert sum(v == 1 for v in products) == 3
134
+ assert sum(sp.simplify(v-(12-3/t**2)) == 0 for v in products) == 1
135
+ rational_coeff = {"".join(map(str, sorted(S))): str(value.subs(t, 2))
136
+ for S, value in coefficient.items()}
137
+ integer_coeff = {key: int(4*sp.Rational(value)) for key, value in rational_coeff.items()}
138
+ integer_products = [int(16*v.subs(t, 2)) for v in products]
139
+ assert sorted(integer_products) == [16, 16, 16, 180]
140
+ assert 180 > (3*4)**2
141
+
142
+ # Direct independent rational-matrix certificate at t=2: alternating signs
143
+ # of every nonempty principal minor are equivalent to <=1 positive eigenvalue.
144
+ rational_minor_count = 0
145
+ for cert in certificates:
146
+ H = sp.Matrix([[sp.sympify(v, locals={"t": t}).subs(t, 2) for v in row] for row in cert["hessian"]])
147
+ for size in range(1, 6):
148
+ for labels in combinations(range(5), size):
149
+ det = sp.factor(H.extract(labels, labels).det())
150
+ assert (-1)**size*det <= 0
151
+ rational_minor_count += 1
152
+
153
+ # Strict interior family. Every reduced derivative Hessian now has
154
+ # signature (1,4), rather than a zero eigenvalue in two symmetry classes.
155
+ delta = sp.symbols("delta", real=True, positive=True)
156
+ strict_coefficient = dict(coefficient)
157
+ strict_coefficient[frozenset(P | {distinguished})] = 3-delta
158
+ strict_coefficient[frozenset(Q)] = 4-t**-2-delta
159
+ strict_certificates = []
160
+ strict_minor_count = 0
161
+ strict_integer_hessian_determinants = []
162
+ strict_integer_hessian_row_bounds = []
163
+ for cert in certificates:
164
+ k = cert["derivative_label"]
165
+ remaining = cert["remaining_labels"]
166
+ groups = cert["symmetry_groups"]
167
+ H = sp.zeros(5)
168
+ for i, j in combinations(range(5), 2):
169
+ H[i, j] = H[j, i] = strict_coefficient[frozenset((k, remaining[i], remaining[j]))]
170
+ B = sp.zeros(5, len(groups))
171
+ for col, group in enumerate(groups):
172
+ for label in group:
173
+ B[remaining.index(label), col] = 1/sp.sqrt(len(group))
174
+ quotient = sp.simplify(B.T*H*B)
175
+ assert sp.simplify(H*B-B*quotient) == sp.zeros(5, len(groups))
176
+ complete_basis = [B[:, j] for j in range(B.cols)]
177
+ for group in groups:
178
+ if len(group) < 2:
179
+ continue
180
+ eigenvalue = -H[remaining.index(group[0]), remaining.index(group[1])]
181
+ for label in group[1:]:
182
+ v = sp.zeros(5, 1)
183
+ v[remaining.index(group[0])] = 1
184
+ v[remaining.index(label)] = -1
185
+ assert sp.simplify(H*v-eigenvalue*v) == sp.zeros(5, 1)
186
+ complete_basis.append(v)
187
+ assert sp.Matrix.hstack(*complete_basis).det() != 0
188
+ determinant = sp.factor(quotient.det())
189
+ second = e2(quotient)
190
+ if k == distinguished:
191
+ assert sp.simplify(determinant-(6-2*delta-6*t**2)) == 0
192
+ reason = "determinant=-6(t^2-1)-2delta<0; two quotient eigenvalues have opposite signs"
193
+ modes = [{"eigenvalue": "delta-3", "multiplicity": 1},
194
+ {"eigenvalue": "-1", "multiplicity": 2}]
195
+ elif k in P:
196
+ assert sp.simplify(determinant-2*(3-delta)*t*delta) == 0
197
+ assert sp.simplify(second-(-(3-delta)**2-3-3*t**2)) == 0
198
+ reason = "determinant=2(3-delta)t delta>0 and e2=-(3-delta)^2-3-3t^2<0; one positive and two negative quotient eigenvalues"
199
+ modes = [{"eigenvalue": "-t", "multiplicity": 2}]
200
+ else:
201
+ assert sp.simplify(determinant-2*t**2*delta) == 0
202
+ assert sp.simplify(second-(2-t**-2-delta-6*t**2)) == 0
203
+ reason = "determinant=2t^2delta>0 and e2=2-t^(-2)-delta-6t^2<0; one positive and two negative quotient eigenvalues"
204
+ modes = [{"eigenvalue": "-1", "multiplicity": 1},
205
+ {"eigenvalue": "delta-4+t^(-2)", "multiplicity": 1}]
206
+ strict_certificates.append({
207
+ "derivative_label": k,
208
+ "orthonormal_quotient": exact_matrix(quotient),
209
+ "quotient_determinant": str(determinant),
210
+ "quotient_sum_principal_2_minors": str(second),
211
+ "strict_sign_certificate": reason,
212
+ "orthogonal_modes": modes,
213
+ "reduced_hessian_inertia_positive_negative_zero": [1, 4, 0],
214
+ })
215
+ H0 = H.subs({t: 2, delta: sp.Rational(1, 4)})
216
+ integer_H0 = 4*H0
217
+ strict_integer_hessian_determinants.append(int(integer_H0.det()))
218
+ strict_integer_hessian_row_bounds.append(
219
+ int(max(sum(abs(integer_H0[i, j]) for j in range(5)) for i in range(5))))
220
+ for size in range(1, 6):
221
+ for labels in combinations(range(5), size):
222
+ det = sp.factor(H0.extract(labels, labels).det())
223
+ if size == 1:
224
+ assert det == 0
225
+ else:
226
+ assert (-1)**size*det < 0
227
+ strict_minor_count += 1
228
+ strict_integer_coeff = {
229
+ "".join(map(str, sorted(S))): int(4*value.subs({t: 2, delta: sp.Rational(1, 4)}))
230
+ for S, value in strict_coefficient.items()}
231
+ strict_products = [sp.factor(strict_coefficient[I | {j}]*strict_coefficient[J-{j}])
232
+ for j in sorted(J)]
233
+ strict_integer_products = [int(16*v.subs({t: 2, delta: sp.Rational(1, 4)}))
234
+ for v in strict_products]
235
+ assert sorted(strict_integer_products) == [16, 16, 16, 154]
236
+ assert set(strict_integer_coeff.values()) == {4, 8, 11, 14}
237
+ assert min(strict_integer_hessian_determinants) == 7168
238
+ assert max(strict_integer_hessian_row_bounds) == 35
239
+ radius = sp.Rational(1, 1000)
240
+ eigenvalue_absolute_lower_bound = sp.Rational(7168, 35**4)
241
+ assert 4*radius < eigenvalue_absolute_lower_bound
242
+ worst_case_squared_failure = (11-radius)*(14-radius)-9*(4+radius)**2
243
+ assert worst_case_squared_failure > 0
244
+
245
+ result = {
246
+ "title": "Exact Lorentzian obstruction to strong T_2",
247
+ "parameter_domain": "real t>1",
248
+ "support": "U_(3,6), all twenty coefficients positive",
249
+ "coefficient_classes": {
250
+ "P": sorted(P), "Q": sorted(Q), "singleton": distinguished,
251
+ "PPs": "3", "QQQ": "4-t^(-2)", "PPQ": "1",
252
+ "sQQ": "1", "PsQ": "t", "PQQ": "t"},
253
+ "derivative_certificates": certificates,
254
+ "strong_relation": {"I": sorted(I), "J": sorted(J),
255
+ "products": list(map(str, products)),
256
+ "necessary_parameter": "p >= log_3(12-3/t^2)",
257
+ "uniform_lower_bound": "p >= log_3(12)"},
258
+ "integer_example": {"parameter_t": 2, "global_coefficient_scale": 4,
259
+ "coefficients": integer_coeff,
260
+ "strong_relation_products": integer_products,
261
+ "exact_squared_T2_excess": 36,
262
+ "rational_principal_minors_checked": rational_minor_count},
263
+ "strict_interior_family": {
264
+ "domain": "t>1 and 0<delta<1",
265
+ "coefficients": {"PPs": "3-delta", "QQQ": "4-t^(-2)-delta",
266
+ "PPQ": "1", "sQQ": "1", "PsQ": "t", "PQQ": "t"},
267
+ "derivative_certificates": strict_certificates,
268
+ "strong_relation_products": list(map(str, strict_products)),
269
+ "necessary_parameter": "p >= log_3((3-delta)(4-t^(-2)-delta)) whenever the product exceeds 1",
270
+ "limiting_uniform_lower_bound": "p >= log_3(12)",
271
+ "strict_integer_example": {
272
+ "parameter_t": 2, "parameter_delta": "1/4", "global_coefficient_scale": 4,
273
+ "coefficients": strict_integer_coeff,
274
+ "strong_relation_products": strict_integer_products,
275
+ "exact_squared_T2_excess": 10,
276
+ "rational_principal_minors_checked": strict_minor_count,
277
+ "qualification": "All six reduced derivative Hessians have signature (1,4).",
278
+ "explicit_absolute_coefficient_perturbation_radius": str(radius),
279
+ "hessian_determinants": strict_integer_hessian_determinants,
280
+ "hessian_maximum_absolute_row_sums": strict_integer_hessian_row_bounds,
281
+ "minimum_eigenvalue_absolute_lower_bound": str(eigenvalue_absolute_lower_bound),
282
+ "perturbation_operator_norm_upper_bound": str(4*radius),
283
+ "worst_case_squared_T2_failure_margin": str(worst_case_squared_failure)}},
284
+ "scope": "This lower bound does not establish a matching upper bound or resolve finiteness of the strong parameter.",
285
+ "status": "all exact symbolic and rational checks passed",
286
+ }
287
+ args.output.write_text(json.dumps(result, indent=2)+"\n")
288
+ print(json.dumps({"output": str(args.output.resolve()), "status": result["status"],
289
+ "derivatives": 12,
290
+ "rational_principal_minors": rational_minor_count+strict_minor_count}, indent=2))
291
+
292
+
293
+ if __name__ == "__main__":
294
+ main()
code/verify_strengthened_aggregation.py ADDED
@@ -0,0 +1,389 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ #!/usr/bin/env python3
2
+ """Finite certificates for strengthened hyperfield aggregation theorems.
3
+
4
+ Exact rational, exact algebraic, formal-polynomial, and numerical checks are
5
+ distinguished in the output. These computations supplement the written proofs;
6
+ they do not establish any universal assertion by numerical extrapolation.
7
+ Requires Python 3 and mpmath 1.3.0 for the explicitly numerical checks.
8
+ """
9
+ from __future__ import annotations
10
+
11
+ import argparse
12
+ from collections import Counter
13
+ from fractions import Fraction as Q
14
+ from functools import lru_cache
15
+ from itertools import combinations, combinations_with_replacement, product
16
+ import json
17
+ from math import isqrt
18
+ from pathlib import Path
19
+ import random
20
+
21
+
22
+ def rational_string(x):
23
+ x = Q(x)
24
+ return str(x.numerator) if x.denominator == 1 else f"{x.numerator}/{x.denominator}"
25
+
26
+
27
+ def multiply(a, b):
28
+ return [[sum(a[i][k] * b[k][j] for k in range(len(b)))
29
+ for j in range(len(b[0]))] for i in range(len(a))]
30
+
31
+
32
+ def transpose(a):
33
+ return [list(row) for row in zip(*a)]
34
+
35
+
36
+ def congruence(a, l):
37
+ return multiply(transpose(l), multiply(a, l))
38
+
39
+
40
+ def matrix_equal(a, b):
41
+ return a == b
42
+
43
+
44
+ def products_for_quartet(a, quartet):
45
+ i, j, k, l = quartet
46
+ return [a[i][j] * a[k][l], a[i][k] * a[j][l], a[i][l] * a[j][k]]
47
+
48
+
49
+ def all_matrix_quartets(a):
50
+ for quartet in combinations_with_replacement(range(len(a)), 4):
51
+ yield quartet, products_for_quartet(a, quartet)
52
+
53
+
54
+ def check_matrix_exact(a, parameter):
55
+ assert parameter in (1, 2)
56
+ count = 0
57
+ smallest = None
58
+ for quartet, terms in all_matrix_quartets(a):
59
+ if parameter == 1:
60
+ margin = sum(terms) - 2 * max(terms)
61
+ else:
62
+ x, y, z = terms
63
+ margin = 2 * (x*y + x*z + y*z) - x*x - y*y - z*z
64
+ assert margin >= 0, (parameter, quartet, terms, margin)
65
+ smallest = margin if smallest is None else min(smallest, margin)
66
+ count += 1
67
+ return dict(quartet_count=count,
68
+ minimum_exact_polynomial_margin=rational_string(smallest))
69
+
70
+
71
+ def determinant_magnitudes(columns):
72
+ return [[abs(u[0]*v[1] - u[1]*v[0]) for v in columns] for u in columns]
73
+
74
+
75
+ def rational_matrices():
76
+ rng = random.Random(186741)
77
+ cases = []
78
+ sources = [
79
+ [(1, 0), (0, 1), (1, 1), (1, -1), (0, 0), (1, 0)],
80
+ [(1, 0), (0, 1), (2, 1), (3, 2), (-2, 3), (1, 0)],
81
+ [(0, 0), (1, 2), (2, 4), (1, -1), (3, 1), (0, 1)],
82
+ ]
83
+ for trial in range(8):
84
+ columns = sources[trial % len(sources)]
85
+ l1 = [[Q(rng.choice([0, 0, 1, 2, 3]), rng.choice([1, 2, 3]))
86
+ for _ in range(4)] for _ in range(6)]
87
+ l2 = [[Q(rng.choice([0, 1, 2, 3]), rng.choice([1, 2]))
88
+ for _ in range(3)] for _ in range(4)]
89
+ if trial == 0:
90
+ for row in l1:
91
+ row[-1] = Q(0)
92
+ cases.append((columns, l1, l2))
93
+ return cases
94
+
95
+
96
+ def exact_congruence_checks(cases):
97
+ out = []
98
+ for trial, (columns, l1, l2) in enumerate(cases, 1):
99
+ minors = determinant_magnitudes(columns)
100
+ for q in (1, 2):
101
+ a = [[Q(x)**q for x in row] for row in minors]
102
+ source = check_matrix_exact(a, q)
103
+ h1 = congruence(a, l1)
104
+ h2 = congruence(h1, l2)
105
+ direct = congruence(a, multiply(l1, l2))
106
+ assert matrix_equal(h2, direct)
107
+ assert all(x >= 0 for row in h2 for x in row)
108
+ out.append(dict(trial=trial, input_parameter=q,
109
+ source_check=source,
110
+ first_substitution=check_matrix_exact(h1, q),
111
+ composed_substitution=check_matrix_exact(h2, q),
112
+ nonzero_diagonal_after_first=sum(h1[i][i] > 0 for i in range(4)),
113
+ exact_composition_identity=True,
114
+ arithmetic="exact rational"))
115
+ return out
116
+
117
+
118
+ def polynomial_add(a, b):
119
+ out = Counter(a)
120
+ out.update(b)
121
+ return {k: v for k, v in out.items() if v}
122
+
123
+
124
+ def polynomial_multiply(a, b):
125
+ out = Counter()
126
+ for m, c in a.items():
127
+ for n, d in b.items():
128
+ out[tuple(sorted(m+n))] += c*d
129
+ return dict(out)
130
+
131
+
132
+ def formal_tensor_factorizations():
133
+ n = 3
134
+ def asym(i, j):
135
+ return "A" + str(min(i, j)) + str(max(i, j))
136
+ def entry(a, b):
137
+ out = Counter()
138
+ for i, j in product(range(n), repeat=2):
139
+ out[tuple(sorted((asym(i, j), f"L{i}{a}", f"L{j}{b}")))] += 1
140
+ return dict(out)
141
+ pairings = [((0, 1), (2, 3)), ((0, 2), (1, 3)), ((0, 3), (1, 2))]
142
+ out = []
143
+ for pairing in pairings:
144
+ (a, b), (c, d) = pairing
145
+ lhs = polynomial_multiply(entry(a, b), entry(c, d))
146
+ rhs = Counter()
147
+ for indices in product(range(n), repeat=4):
148
+ i, j, k, l = indices
149
+ rhs[tuple(sorted((asym(indices[a], indices[b]),
150
+ asym(indices[c], indices[d]),
151
+ f"L{i}0", f"L{j}1", f"L{k}2", f"L{l}3")))] += 1
152
+ assert lhs == dict(rhs)
153
+ out.append(dict(pairing=pairing, expanded_monomials=len(lhs),
154
+ total_summands=sum(lhs.values()), status="identity verified"))
155
+ return dict(arithmetic="exact commutative polynomial identities",
156
+ source_indices=n, target_indices=4, factorizations=out)
157
+
158
+
159
+ def numerical_congruence_checks(cases):
160
+ import mpmath as mp
161
+ mp.mp.dps = 90
162
+ def m(x):
163
+ x = Q(x)
164
+ return mp.mpf(x.numerator) / x.denominator
165
+ samples = ["0.25", "0.5", "1", "1.5", "2", "3", "4", "8"]
166
+ out = []
167
+ for trial, (columns, l1, l2) in enumerate(cases, 1):
168
+ minors = determinant_magnitudes(columns)
169
+ ml1 = [[m(x) for x in row] for row in l1]
170
+ ml2 = [[m(x) for x in row] for row in l2]
171
+ for qs in samples:
172
+ q = mp.mpf(qs)
173
+ p = max(q, mp.mpf(1))
174
+ a = [[m(x)**q for x in row] for row in minors]
175
+ h1 = congruence(a, ml1)
176
+ h2 = congruence(h1, ml2)
177
+ hd = congruence(a, multiply(ml1, ml2))
178
+ for row1, row2 in zip(h2, hd):
179
+ for x, y in zip(row1, row2):
180
+ assert abs(x-y) <= mp.mpf("1e-78") * max(abs(x), abs(y), 1)
181
+ for stage, h in (("first", h1), ("composed", h2)):
182
+ count = 0
183
+ smallest = mp.mpf(1)
184
+ for quartet, terms in all_matrix_quartets(h):
185
+ roots = [x**(1/p) for x in terms]
186
+ scale = max(roots)
187
+ rel = mp.mpf(0) if not scale else (sum(roots)-2*scale)/scale
188
+ assert rel >= -mp.mpf("1e-76"), (trial, qs, stage, quartet, rel)
189
+ smallest = min(smallest, rel)
190
+ count += 1
191
+ out.append(dict(trial=trial, input_parameter=qs,
192
+ output_parameter=mp.nstr(p, 15), stage=stage,
193
+ quartet_count=count,
194
+ minimum_relative_root_polygon_slack=mp.nstr(smallest, 30),
195
+ precision_decimal_digits=90,
196
+ arithmetic="high-precision numerical; not an exact certificate"))
197
+ return out
198
+
199
+
200
+ def minimal_threshold_certificate():
201
+ columns = [(1, 0), (0, 1), (1, 64), (64, -1)]
202
+ mm = determinant_magnitudes(columns)
203
+ input_lengths = [mm[0][1]*mm[2][3], mm[0][2]*mm[1][3], mm[0][3]*mm[1][2]]
204
+ assert input_lengths == [4097, 4096, 1]
205
+ assert sum(input_lengths)-2*max(input_lengths) == 0
206
+ assert mm[0][2] == mm[1][3] == 64
207
+ assert mm[0][3] == mm[1][2] == 1
208
+ output_root_lengths = [4*4097, 81**2, 81**2]
209
+ defect = max(output_root_lengths) - sum(output_root_lengths[1:])
210
+ assert output_root_lengths == [16388, 6561, 6561]
211
+ assert defect == 3266
212
+ return dict(input_parameter="1/2", columns=columns,
213
+ input_root_product_lengths=input_lengths,
214
+ normalized_output_products=["2*sqrt(4097)", "81", "81"],
215
+ output_root_product_lengths=output_root_lengths,
216
+ exact_output_root_defect=defect,
217
+ input_support="U_(2,4); all six minors nonzero",
218
+ arithmetic="exact integers and stated algebraic square roots")
219
+
220
+
221
+ def strong_relation_products(coeff, d, n):
222
+ for raw_i in combinations(range(n), d-1):
223
+ i = frozenset(raw_i)
224
+ for raw_j in combinations(range(n), d+1):
225
+ j = frozenset(raw_j)
226
+ yield i, j, [coeff[i | {a}]*coeff[j - {a}] for a in sorted(j-i)]
227
+
228
+
229
+ def weak_relation_products(coeff, d, n):
230
+ for cc in combinations(range(n), d-2):
231
+ c = frozenset(cc)
232
+ for i, j, k, l in combinations(sorted(set(range(n))-c), 4):
233
+ yield c, (i, j, k, l), [coeff[c | {i, j}]*coeff[c | {k, l}],
234
+ coeff[c | {i, k}]*coeff[c | {j, l}],
235
+ coeff[c | {i, l}]*coeff[c | {j, k}]]
236
+
237
+
238
+ def two_peak_coefficients(d, n, peak1, peak2, default):
239
+ c = set(range(6, d+3))
240
+ u, v = frozenset(c | {0, 2, 4}), frozenset(c | {1, 3, 5})
241
+ coeff = {frozenset(s): Q(default) for s in combinations(range(n), d)}
242
+ coeff[u], coeff[v] = Q(peak1), Q(peak2)
243
+ return coeff, c
244
+
245
+
246
+ def two_peak_weak_checks():
247
+ out = []
248
+ for d, n in ((3, 6), (4, 7), (5, 8), (4, 8)):
249
+ coeff, _ = two_peak_coefficients(d, n, 2, 2, 1)
250
+ count = 0
251
+ types = Counter()
252
+ for _, _, terms in weak_relation_products(coeff, d, n):
253
+ assert sum(terms)-2*max(terms) >= 0
254
+ types[tuple(sorted(terms))] += 1
255
+ count += 1
256
+ strong_failures = 0
257
+ for _, _, terms in strong_relation_products(coeff, d, n):
258
+ strong_failures += sum(terms)-2*max(terms) < 0
259
+ assert strong_failures > 0
260
+ out.append(dict(rank=d, ground_size=n, weak_relations=count,
261
+ root_product_patterns={str(tuple(rational_string(t) for t in k)): v
262
+ for k, v in types.items()},
263
+ failed_strong_relations=strong_failures,
264
+ arithmetic="exact rational root products"))
265
+ return dict(instances=out,
266
+ universal_output_products=["(1+2^q)^2", "4", "4"],
267
+ necessary_output_parameter="p >= 2*log_2((1+2^q)/2)",
268
+ qualification="The universal formula requires the written symbolic proof.")
269
+
270
+
271
+ @lru_cache(maxsize=None)
272
+ def sqrt_interval(x, digits=80):
273
+ x = Q(x)
274
+ scale = 10**digits
275
+ k = isqrt((x.numerator*scale*scale)//x.denominator)
276
+ lo = Q(k, scale)
277
+ hi = lo if lo*lo == x else Q(k+1, scale)
278
+ assert lo*lo <= x <= hi*hi
279
+ return lo, hi
280
+
281
+
282
+ def two_peak_strong_checks():
283
+ out = []
284
+ for d, n in ((3, 6), (4, 7), (5, 8), (4, 8), (5, 9)):
285
+ roots, _ = two_peak_coefficients(d, n, 2, Q(3, 2), 1)
286
+ weak_count = genuine_count = taut_count = 0
287
+ root_patterns = Counter()
288
+ for _, _, terms in strong_relation_products(roots, d, n):
289
+ assert sum(terms)-2*max(terms) >= 0
290
+ root_patterns[tuple(sorted(terms))] += 1
291
+ if len(terms) == 2:
292
+ taut_count += 1
293
+ else:
294
+ genuine_count += 1
295
+ for _, _, terms in weak_relation_products(roots, d, n):
296
+ assert sum(terms)-2*max(terms) >= 0
297
+ weak_count += 1
298
+ integer, c = two_peak_coefficients(d, n, 23, 14, 6)
299
+ smallest = None
300
+ strict_count = 0
301
+ for i, j, terms in strong_relation_products(integer, d, n):
302
+ if len(terms) == 2:
303
+ assert terms[0] == terms[1]
304
+ continue
305
+ largest = max(range(len(terms)), key=terms.__getitem__)
306
+ intervals = [sqrt_interval(t) for t in terms]
307
+ certified_margin = sum(lo for k, (lo, hi) in enumerate(intervals)
308
+ if k != largest) - intervals[largest][1]
309
+ assert certified_margin > 0, (d, n, i, j, terms)
310
+ smallest = certified_margin if smallest is None else min(smallest, certified_margin)
311
+ strict_count += 1
312
+ g = {str(i+1)+str(j+1): integer[frozenset(c | {i, j, 4})]
313
+ + integer[frozenset(c | {i, j, 5})]
314
+ for i, j in combinations(range(4), 2)}
315
+ assert g["13"] == 29 and g["24"] == 20
316
+ assert all(g[t] == 12 for t in ("12", "14", "23", "34"))
317
+ products_out = [g["13"]*g["24"], g["12"]*g["34"], g["14"]*g["23"]]
318
+ assert products_out == [580, 144, 144]
319
+ squared_excess = products_out[0] - 24**2
320
+ assert squared_excess == 4
321
+ out.append(dict(rank=d, ground_size=n, weak_relation_count=weak_count,
322
+ strong_genuine_relation_count=genuine_count,
323
+ strong_tautological_relation_count=taut_count,
324
+ strong_root_peak_values=["2", "3/2"],
325
+ strict_integer_coefficient_values=[6, 14, 23],
326
+ strict_integer_input_relations_certified=strict_count,
327
+ exact_minimum_root_margin_lower_bound=rational_string(smallest),
328
+ output_coefficients={k: rational_string(v) for k, v in g.items()},
329
+ output_products=[int(x) for x in products_out],
330
+ exact_squared_output_excess=int(squared_excess),
331
+ arithmetic="exact rational root products and rational square-root intervals"))
332
+ return dict(instances=out,
333
+ universal_two_peak_output_products=["(1+2^q)*(1+(3/2)^q)", "4", "4"],
334
+ necessary_output_parameter="p >= log_2(((1+2^q)*(1+(3/2)^q))/4)",
335
+ exact_boundary_q2_output_products=["65/4", "4", "4"],
336
+ boundary_q2_necessary_output_parameter="p >= log_2(65/16) > 2",
337
+ qualification="Universal validity requires the written all-rank proof.")
338
+
339
+
340
+ def source_size_relation_counts():
341
+ out = []
342
+ for d in range(2, 7):
343
+ for corank in (0, 1, 2, 3):
344
+ n = d+corank
345
+ counts = Counter()
346
+ for i in combinations(range(n), d-1):
347
+ for j in combinations(range(n), d+1):
348
+ counts[len(set(j)-set(i))] += 1
349
+ largest = max(counts, default=0)
350
+ assert largest <= corank+1
351
+ if corank <= 1:
352
+ assert largest <= 2
353
+ if corank <= 2:
354
+ assert largest <= 3
355
+ out.append(dict(rank=d, ground_size=n, corank=corank,
356
+ strong_term_counts=dict(sorted(counts.items())),
357
+ maximum_summands=largest))
358
+ return dict(arithmetic="exact finite combinatorial enumeration", instances=out)
359
+
360
+
361
+ def main():
362
+ parser = argparse.ArgumentParser()
363
+ parser.add_argument("--output", type=Path,
364
+ default=Path(__file__).with_name("strengthened_verification_results.json"))
365
+ args = parser.parse_args()
366
+ cases = rational_matrices()
367
+ result = dict(
368
+ title="Strengthened aggregation reproducibility certificates",
369
+ scope="Finite checks supplement the self-contained mathematical proofs.",
370
+ exact_quadratic_congruence=exact_congruence_checks(cases),
371
+ formal_tensor_factorizations=formal_tensor_factorizations(),
372
+ numerical_quadratic_congruence=numerical_congruence_checks(cases),
373
+ exact_minimal_threshold=minimal_threshold_certificate(),
374
+ exact_two_peak_weak=two_peak_weak_checks(),
375
+ exact_two_peak_strong=two_peak_strong_checks(),
376
+ exact_corank_relation_counts=source_size_relation_counts(),
377
+ )
378
+ args.output.parent.mkdir(parents=True, exist_ok=True)
379
+ args.output.write_text(json.dumps(result, indent=2)+"\n")
380
+ print(json.dumps(dict(output=str(args.output.resolve()),
381
+ exact_congruence_instances=len(result["exact_quadratic_congruence"]),
382
+ numerical_congruence_instances=len(result["numerical_quadratic_congruence"]),
383
+ weak_two_peak_instances=len(result["exact_two_peak_weak"]["instances"]),
384
+ strong_two_peak_instances=len(result["exact_two_peak_strong"]["instances"]),
385
+ status="all checks passed"), indent=2))
386
+
387
+
388
+ if __name__ == "__main__":
389
+ main()
code/verify_strong_parameter_bound.py ADDED
@@ -0,0 +1,145 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ #!/usr/bin/env python3
2
+ """Exact strong T_2 inputs whose merge fails T_(20/9).
3
+
4
+ Only standard-library rational arithmetic is used. The universal all-rank
5
+ extension is a written theorem; this executable certifies the rank-three
6
+ integer witnesses and all 225 indexed strong input relations.
7
+ """
8
+ from fractions import Fraction
9
+ from itertools import combinations
10
+ from math import isqrt
11
+ from pathlib import Path
12
+ import json
13
+
14
+
15
+ def rational(value):
16
+ value = Fraction(value)
17
+ return {"numerator": value.numerator, "denominator": value.denominator}
18
+
19
+
20
+ def sqrt_interval(value, digits=80):
21
+ """Rational endpoints proved to contain sqrt(value) by squaring."""
22
+ value = Fraction(value)
23
+ scale = 10 ** digits
24
+ integer = isqrt(value.numerator * scale * scale // value.denominator)
25
+ lower = Fraction(integer, scale)
26
+ upper = lower if lower * lower == value else Fraction(integer + 1, scale)
27
+ assert lower * lower <= value <= upper * upper
28
+ return lower, upper
29
+
30
+
31
+ def certify(scale, small_coefficient):
32
+ coefficients = {}
33
+ for subset in combinations(range(1, 7), 3):
34
+ key = frozenset(subset)
35
+ if key == frozenset((1, 3, 5)):
36
+ value = 4 * scale - 1
37
+ elif key in (frozenset((1, 2, 6)), frozenset((3, 4, 6))):
38
+ value = small_coefficient
39
+ elif 6 in key:
40
+ value = 5 * scale
41
+ else:
42
+ value = scale
43
+ coefficients[key] = value
44
+
45
+ strict_count = tautological_count = 0
46
+ minimum_slack_bound = None
47
+ relation_certificates = []
48
+ for I_tuple in combinations(range(1, 7), 2):
49
+ I = frozenset(I_tuple)
50
+ for J_tuple in combinations(range(1, 7), 4):
51
+ J = frozenset(J_tuple)
52
+ values = [coefficients[I | {j}] * coefficients[J - {j}]
53
+ for j in sorted(J - I)]
54
+ if len(values) == 2:
55
+ assert values[0] == values[1]
56
+ tautological_count += 1
57
+ else:
58
+ index = max(range(len(values)), key=values.__getitem__)
59
+ intervals = [sqrt_interval(value) for value in values]
60
+ lower_slack = sum(intervals[j][0] for j in range(len(values))
61
+ if j != index) - intervals[index][1]
62
+ assert lower_slack > 0
63
+ strict_count += 1
64
+ if minimum_slack_bound is None or lower_slack < minimum_slack_bound:
65
+ minimum_slack_bound = lower_slack
66
+ relation_certificates.append({
67
+ "I": list(I_tuple), "J": list(J_tuple),
68
+ "products": values,
69
+ "exact_root_intervals": [[rational(lo), rational(hi)]
70
+ for lo, hi in intervals],
71
+ "positive_polygon_slack_lower_bound": rational(lower_slack),
72
+ })
73
+ assert strict_count == 135 and tautological_count == 90
74
+
75
+ merged = {}
76
+ for i, j in combinations(range(1, 5), 2):
77
+ merged[f"{i}{j}"] = (coefficients[frozenset((i, j, 5))]
78
+ + coefficients[frozenset((i, j, 6))])
79
+ A = Fraction(merged["13"] * merged["24"], scale ** 2)
80
+ B = Fraction(merged["14"] * merged["23"], scale ** 2)
81
+ C = Fraction(merged["12"] * merged["34"], scale ** 2)
82
+ left_lower = Fraction(6019, 1000)
83
+ first_right_upper = Fraction(5016, 1000)
84
+ second_right_upper = Fraction(1001, 1000)
85
+
86
+ # Since 1/(20/9)=9/20, each interval claim becomes an integer-power
87
+ # comparison after raising to the positive twentieth power.
88
+ differences = [A ** 9 - left_lower ** 20,
89
+ first_right_upper ** 20 - B ** 9,
90
+ second_right_upper ** 20 - C ** 9]
91
+ assert all(difference > 0 for difference in differences)
92
+ output_defect_bound = left_lower - first_right_upper - second_right_upper
93
+ assert output_defect_bound == Fraction(1, 500)
94
+
95
+ # The baseline has peak r<4 and small coefficient epsilon<=1/4;
96
+ # diagonal weight 5 on element 6 and global scale N give this array.
97
+ peak = Fraction(4 * scale - 1, scale)
98
+ epsilon = Fraction(small_coefficient, 5 * scale)
99
+ assert 1 < peak < 4 and 0 < epsilon <= Fraction(1, 4)
100
+ assert peak * epsilon <= 1
101
+
102
+ return {
103
+ "arithmetic": "exact rational arithmetic and certified rational root intervals",
104
+ "input_parameter": rational(2),
105
+ "failed_output_parameter": rational(Fraction(20, 9)),
106
+ "global_coefficient_scale": scale,
107
+ "element6_coefficient_weight": 5,
108
+ "baseline_peak": rational(peak),
109
+ "baseline_small_coefficient": rational(epsilon),
110
+ "input_coefficients": {"".join(map(str, sorted(k))): v
111
+ for k, v in coefficients.items()},
112
+ "strict_input_relations": strict_count,
113
+ "tautological_input_relations": tautological_count,
114
+ "minimum_input_root_polygon_slack_lower_bound": rational(minimum_slack_bound),
115
+ "input_relation_certificates": relation_certificates,
116
+ "merged_coefficients": merged,
117
+ "normalized_output_products": {"A": rational(A), "B": rational(B),
118
+ "C": rational(C)},
119
+ "root_exponent": rational(Fraction(9, 20)),
120
+ "left_root_lower_bound": rational(left_lower),
121
+ "first_right_root_upper_bound": rational(first_right_upper),
122
+ "second_right_root_upper_bound": rational(second_right_upper),
123
+ "positive_twentieth_power_comparison_differences": [rational(d)
124
+ for d in differences],
125
+ "normalized_output_defect_strictly_greater_than": rational(output_defect_bound),
126
+ "status": "all certificates passed",
127
+ }
128
+
129
+
130
+ def main():
131
+ results = {
132
+ "theorem_consequence": "The universal strong-input repair at q=2 is strictly greater than 20/9.",
133
+ "qualification": "The strict inequality for the universal infimum follows by continuity in the output parameter, using this fixed strict witness.",
134
+ "witnesses": [certify(100000, 5), certify(1000, 1)],
135
+ }
136
+ output = Path(__file__).with_name("strong_parameter_bound_results.json")
137
+ output.write_text(json.dumps(results, indent=2) + "\n")
138
+ print(json.dumps({"output": str(output.resolve()), "witnesses": 2,
139
+ "indexed_input_relations_per_witness": 225,
140
+ "arithmetic": "exact only", "status": "all checks passed"},
141
+ indent=2))
142
+
143
+
144
+ if __name__ == "__main__":
145
+ main()
data/examples.jsonl ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {"id": "example-three-level-strict-q2", "record_type": "exact_example", "title": "Strict integer levels 6, 14, 23", "content": "On U_(3,6), coefficients 23 at 135, 14 at 246, and 6 elsewhere are strict strong T_2. Merging labels 5 and 6 yields products (580,144,144); 580-(12+12)^2=4>0.", "source_section": "8", "verification_method": "exact_integer_certificate", "verification_status": "checked_exactly", "artifact_paths": ["paper/Ilyoxune_Aggregation_Theorem.pdf", "paper/Ilyoxune_Aggregation_Theorem.tex", "code/verify_strengthened_aggregation.py", "results/strengthened_verification_results.json"], "payload_json": "{\"coefficients\": {\"123\": 6, \"124\": 6, \"125\": 6, \"126\": 6, \"134\": 6, \"135\": 23, \"136\": 6, \"145\": 6, \"146\": 6, \"156\": 6, \"234\": 6, \"235\": 6, \"236\": 6, \"245\": 6, \"246\": 14, \"256\": 6, \"345\": 6, \"346\": 6, \"356\": 6, \"456\": 6}, \"ground_set_size\": 6, \"merged_labels\": [5, 6], \"output_products\": [580, 144, 144], \"q\": \"2\", \"rank\": 3, \"squared_excess\": 4}"}
2
+ {"id": "example-strong-parameter-20-over-9", "record_type": "exact_example", "title": "Strong-input repair barrier above 20/9", "content": "The strict uniform T_2 input with coefficient levels 1,1000,3999,5000 fails normalized weak T_(20/9) after one merge. The root comparisons are certified with integer twentieth powers.", "source_section": "8", "verification_method": "exact_integer_and_rational_certificate", "verification_status": "checked_exactly", "artifact_paths": ["paper/Ilyoxune_Aggregation_Theorem.pdf", "paper/Ilyoxune_Aggregation_Theorem.tex", "code/verify_strong_parameter_bound.py", "results/strong_parameter_bound_results.json"], "payload_json": "{\"coefficients\": {\"123\": 1000, \"124\": 1000, \"125\": 1000, \"126\": 1, \"134\": 1000, \"135\": 3999, \"136\": 5000, \"145\": 1000, \"146\": 5000, \"156\": 5000, \"234\": 1000, \"235\": 1000, \"236\": 5000, \"245\": 1000, \"246\": 5000, \"256\": 5000, \"345\": 1000, \"346\": 1, \"356\": 5000, \"456\": 5000}, \"ground_set_size\": 6, \"merged_coefficients\": {\"12\": 1001, \"13\": 8999, \"14\": 6000, \"23\": 6000, \"24\": 6000, \"34\": 1001}, \"merged_labels\": [5, 6], \"normalized_product_values\": [\"26997/500\", \"36\", \"1002001/1000000\"], \"output_parameter\": \"20/9\", \"q\": \"2\", \"rank\": 3, \"root_intervals\": {\"largest_lower\": \"6019/1000\", \"second_upper\": \"5016/1000\", \"third_upper\": \"1001/1000\"}}"}
3
+ {"id": "example-minimal-corank-q-half", "record_type": "exact_example", "title": "Smallest nontrivial one-merge failure below one", "content": "At q=1/2, columns (1,0),(0,1),(1,64),(64,-1) give input root products (4097,4096,1). Merging the last two gives output root products (16388,6561,6561), with excess 3266.", "source_section": "7", "verification_method": "exact_algebraic_and_integer_certificate", "verification_status": "checked_exactly", "artifact_paths": ["paper/Ilyoxune_Aggregation_Theorem.pdf", "paper/Ilyoxune_Aggregation_Theorem.tex", "code/verify_strengthened_aggregation.py", "results/strengthened_verification_results.json"], "payload_json": "{\"coefficient_rule\": \"absolute maximal minors raised to q\", \"excess\": 3266, \"ground_set_size\": 4, \"input_root_products\": [4097, 4096, 1], \"matrix\": [[1, 0, 1, 64], [0, 1, 64, -1]], \"output_root_products\": [16388, 6561, 6561], \"q\": \"1/2\", \"rank\": 2}"}
4
+ {"id": "example-lorentzian-integer-open-box", "record_type": "exact_example", "title": "Robust Lorentzian failure of strong T_2", "content": "A positive U_(3,6) cubic with coefficient levels 4,8,11,14 has active derivative Hessian signatures (1,4), yet strong products (154,16,16,16) fail T_2 since 154>144. Every absolute coefficient perturbation of size at most 1/1000 retains both properties.", "source_section": "10", "verification_method": "exact_symbolic_hessian_and_integer_certificate", "verification_status": "checked_exactly", "artifact_paths": ["paper/Ilyoxune_Aggregation_Theorem.pdf", "paper/Ilyoxune_Aggregation_Theorem.tex", "code/verify_lorentzian_obstruction.py", "results/lorentzian_obstruction_results.json"], "payload_json": "{\"absolute_perturbation_radius\": \"1/1000\", \"active_hessian_determinants\": [11264, 7168, 11264, 7168, 52096, 7168], \"coefficients\": {\"123\": 4, \"124\": 8, \"125\": 8, \"126\": 8, \"134\": 4, \"135\": 11, \"136\": 4, \"145\": 8, \"146\": 8, \"156\": 8, \"234\": 8, \"235\": 8, \"236\": 8, \"245\": 4, \"246\": 14, \"256\": 4, \"345\": 8, \"346\": 8, \"356\": 8, \"456\": 4}, \"ground_set_size\": 6, \"rank\": 3, \"squared_excess\": 10, \"strong_products\": [154, 16, 16, 16]}"}
5
+ {"id": "example-strict-rational-q1", "record_type": "exact_example", "title": "Strict rational counterexample at parameter one", "content": "Root peak 199/100 at 135, lows 1/4 at 126 and 346, default one, and label-six scale 1/4 produce a strict strong T_1 input. The merged relation defect is exactly 139/1280.", "source_section": "11", "verification_method": "exact_rational_certificate", "verification_status": "checked_exactly", "artifact_paths": ["paper/Ilyoxune_Aggregation_Theorem.pdf", "paper/Ilyoxune_Aggregation_Theorem.tex", "code/verify_ilyoxune.py", "results/verification_results.json"], "payload_json": "{\"arithmetic\": \"exact rational\", \"element6_scale\": \"1/4\", \"output_coefficients\": {\"12\": \"17/16\", \"13\": \"56/25\", \"14\": \"5/4\", \"23\": \"5/4\", \"24\": \"5/4\", \"34\": \"17/16\"}, \"output_relation_defect\": \"139/1280\", \"q\": 1, \"root_low\": \"1/4\", \"root_peak\": \"199/100\", \"strict_relations\": 135}"}
6
+ {"id": "example-optimal-quadratic-lower-one", "record_type": "exact_example", "title": "Exact quadratic lower bound p >= 1", "content": "Start from F=x_1*x_2 and substitute x_1=y_1+z, x_2=y_2+z. The output y_1*y_2+y_1*z+y_2*z+z^2 has normalized repeated-z quartet products (2,1,1), forcing p>=1.", "source_section": "3", "verification_method": "written_and_exact_integer_certificate", "verification_status": "checked_exactly", "artifact_paths": ["paper/Ilyoxune_Aggregation_Theorem.pdf", "paper/Ilyoxune_Aggregation_Theorem.tex", "code/verify_strengthened_aggregation.py"], "payload_json": "{\"normalization\": \"rho_beta = beta! times raw coefficient\", \"output_products\": [2, 1, 1], \"source_matrix\": [[0, 1], [1, 0]], \"substitution_matrix\": [[1, 0, 1], [0, 1, 1]]}"}
data/manuscript.jsonl ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"id": "section-1", "record_type": "manuscript_section", "title": "Aggregation laws", "content": "\\section{Aggregation laws}\nCoefficient compatibility is governed by coupled quadratic relations. A substitution adds coefficients and may change their global compatibility, even when every original relation holds strictly. We establish exact low-degree preservation laws, a sharp cardinality boundary, and constructive weak repair in every degree.\n\n\\begin{theorem}[Principal results]\\label{thm:main}\nLet $q>0$.\n\\begin{enumerate}\n\\item For any symmetric nonnegative matrix $C$ satisfying all $\\T_q$ quartet inequalities and every nonnegative rectangular matrix $A$, the matrix $A^{\\mathsf T}CA$ satisfies all quartet inequalities at the exact optimal universal parameter $\\max\\{q,1\\}$. This controls the full normalized quadratic polynomial under arbitrary nonnegative linear substitutions.\n\\item One merge of rank-$d$ matroid data on $n$ labels preserves strong representability at the same parameter precisely in the regimes in Table~\\ref{tab:threshold}. The same classification holds for universal strong-to-weak preservation.\n\\item Every weak $\\T_q$ matroid input, including zero coefficients, has full normalized weak $\\T_{2\\max(q,1)}$ output after one merge. The output parameter is exactly optimal for $0<q\\le1$, and its asymptotic factor two is optimal.\n\\item For every $d\\ge3,n\\ge d+3$, explicit strict strong uniform inputs fail after one merge at parameter $q$. At $q=2$ a quantitative certificate forces the universal strong-input repair parameter above $20/9$.\n\\item Weighted set identifications followed by squarefree projection preserve all strong rank-$d$ arrays, $d\\ge3$, exactly when their active map is injective or has at most $d+1$ output labels.\n\\item There is a family in the relative interior of the Lorentzian $U_{3,6}$ stratum forcing every universal Lorentzian-to-strong parameter to be at least $\\log_3 12$.\n\\end{enumerate}\n\\end{theorem}\n\n\\begin{table}[h]\n\\centering\\small\n\\begin{tabular}{lll}\n\\toprule\nDegree & Ground-set size & Same-parameter universal preservation\\\\\\midrule\n$d=1$ & $n\\ge2$ & Every $q>0$\\\\\n$d\\ge2$ & $d\\le n\\le d+1$ & Every $q>0$\\\\\n$d=2$ & $n\\ge4$ & Exactly $q\\ge1$\\\\\n$d\\ge3$ & $n=d+2$ & Exactly $q\\ge1$\\\\\n$d\\ge3$ & $n\\ge d+3$ & No $q>0$\\\\\\bottomrule\n\\end{tabular}\n\\caption{Complete classification for one full normalized merge.}\\label{tab:threshold}\n\\end{table}\n\nA full polynomial and its squarefree projection are distinguished throughout. Normalization is explicit, including repeated monomials. The sparse rank-three pattern behind one obstruction family is credited to Proposition 8.5 of \\cite{BHKL}; every construction used here is proved directly. Historical priority of the new theorem package is not asserted.", "source_section": "1", "verification_method": "standalone_manuscript", "verification_status": "source_text", "artifact_paths": ["paper/Ilyoxune_Aggregation_Theorem.pdf", "paper/Ilyoxune_Aggregation_Theorem.tex"], "payload_json": "{\"content_format\": \"latex\", \"section_number\": \"1\"}"}
2
+ {"id": "section-2", "record_type": "manuscript_section", "title": "Conventions, normalization, and elementary tools", "content": "\\section{Conventions, normalization, and elementary tools}\\label{sec:definitions}\n\\subsection{Strong arrays and weak relations}\nFor $q>0$, a finite list $t_1,\\ldots,t_m\\ge0$ is null over $\\T_q$ when\n\\begin{equation}\\label{eq:null}\n t_k^{1/q}\\le\\sum_{\\ell\\ne k}t_\\ell^{1/q}\\qquad\\text{for every }k.\n\\end{equation}\nA matroid basis family is a nonempty family of equally sized sets satisfying basis exchange: for bases $B,D$ and $i\\in B\\setminus D$, some $j\\in D\\setminus B$ makes $B-i+j$ a basis. For a rank-$d$ coefficient function $c$ on the $d$-subsets of a ground set $E$, strong representability requires the support to be a matroid and requires \\eqref{eq:null} for\n\\begin{equation}\\label{eq:strong}\n \\bigl(c_{I\\cup\\{j\\}}c_{J\\setminus\\{j\\}}:j\\in J\\setminus I\\bigr),\n \\qquad |I|=d-1,\\quad |J|=d+1.\n\\end{equation}\nThe weak relations are the three-term subfamily: for $|K|=d-2$ and four distinct elements $i,j,k,\\ell\\notin K$, the products are\n\\begin{equation}\\label{eq:weak}\n c_{Kij}c_{Kk\\ell},\\qquad c_{Kik}c_{Kj\\ell},\\qquad c_{Ki\\ell}c_{Kjk}.\n\\end{equation}\nConcatenation denotes adjoining distinct indices. All inequalities in this paper are unsigned: the additive inverse of a positive element in this hyperfield equals itself.\n\nIf $m=|J\\setminus I|=2$, the two terms in \\eqref{eq:strong} are identical. ``Strict strong'' always means strictness for $m\\ge3$; it never means making a tautological equality strict.\n\n\\begin{lemma}[Monotonicity and rescaling]\\label{lem:basic}\nIf a list is null at $q$, it is null at every $p\\ge q$. Multiplying coefficients by $\\lambda\\prod_{i\\in S}d_i$, for $\\lambda,d_i>0$, preserves all strong and weak relations and their strictness.\n\\end{lemma}\n\\begin{proof}\nFor monotonicity put $\\theta=q/p\\le1$ and use $(\\sum a_i)^\\theta\\le\\sum a_i^\\theta$ for nonnegative $a_i$. In a fixed strong relation, every product acquires the common factor $\\lambda^2\\prod_{i\\in I}d_i\\prod_{j\\in J}d_j$. The same argument covers weak relations.\n\\end{proof}\n\n\\subsection{What the normalized output means}\nFor a multi-affine input $F=\\sum_{|S|=d}c_Sx^S$, set $G=D_{u,v}F$. Its coefficient function in exponential form is\n\\begin{equation}\\label{eq:rho}\n \\rho_\\beta=\\beta!\\,[y^\\beta]G,\n \\qquad G=\\sum_\\beta\\rho_\\beta\\frac{y^\\beta}{\\beta!}.\n\\end{equation}\nFor squarefree $\\beta$ the factorial equals one. If $A$ avoids $u,v$ and $z$ denotes their merge, then\n\\begin{equation}\\label{eq:output}\n \\rho_A=c_A,\\qquad\n \\rho_{Az}=c_{Au}+c_{Av},\\qquad\n \\rho_{A+2z}=2c_{Auv},\n\\end{equation}\nwhenever the indices have total degree $d$.\n\nA fixed-degree integer set $J$ is M-convex if whenever $x_i>y_i$ for $x,y\\in J$, there is a $j$ with $x_j<y_j$ such that $x-e_i+e_j,y+e_i-e_j\\in J$. For positive uniform input the output support consists of all degree-$d$ exponent vectors with capacities one on unmerged labels and two on $z$. This set is M-convex: if two vectors differ in one coordinate, an opposite coordinate difference supplies a valid exchange without crossing either capacity. Its weak tests are the three products\n\\begin{equation}\\label{eq:generalweak}\n \\rho_{\\alpha+i+j}\\rho_{\\alpha+k+\\ell},\\quad\n \\rho_{\\alpha+i+k}\\rho_{\\alpha+j+\\ell},\\quad\n \\rho_{\\alpha+i+\\ell}\\rho_{\\alpha+j+k},\n\\end{equation}\nwhere $|\\alpha|=d-2$ and $\\alpha+e_i+e_j+e_k+e_\\ell$ respects capacities. The quartet may contain $z$ twice. This is the bounded-multiplicity weak convention of \\cite{BHKL}. A squarefree shadow instead deletes all monomials with an exponent above one.\n\n\\begin{lemma}[Aggregation of triangles]\\label{lem:minkowski}\nLet $X_h,Y_h,Z_h\\ge0$ satisfy all triangle inequalities for each $h$. Put $r=\\max\\{q,1\\}$. Then\n\\[\n (\\sum_hX_h^q)^{1/r},\\quad\n (\\sum_hY_h^q)^{1/r},\\quad\n (\\sum_hZ_h^q)^{1/r}\n\\]\nsatisfy all triangle inequalities.\n\\end{lemma}\n\\begin{proof}\nFor $q\\ge1$ this is Minkowski's inequality, applied to $X_h\\le Y_h+Z_h$, and its cyclic analogues. For $q\\le1$, $X_h^q\\le(Y_h+Z_h)^q\\le Y_h^q+Z_h^q$; sum over $h$.\n\\end{proof}\n\n\\begin{lemma}[Square roots of metrics are Ptolemaic]\\label{lem:snowflake}\nFor every metric $D$ and four points,\n\\begin{equation}\\label{eq:snowflake}\n \\sqrt{D_{13}D_{24}}\\le\n \\sqrt{D_{12}D_{34}}+\\sqrt{D_{14}D_{23}}.\n\\end{equation}\n\\end{lemma}\n\\begin{proof}\nWrite the successive side distances as $a,b,c,e$, and the diagonals as $f,g$. The triangle inequalities give\n$f\\le\\min(a+b,c+e)$ and $g\\le\\min(a+e,b+c)$.\nBy relabeling around the four-cycle the two chosen minima can be written $a+b$ and $a+e$: the two chosen paths always share a side. Their being minima implies $c\\ge a+|b-e|$. It suffices to prove\n\\[\n (a+b)(a+e)\\le(\\sqrt{ac}+\\sqrt{be})^2.\n\\]\nAfter cancellation this is $a(a+b+e-c)\\le2\\sqrt{abce}$. The left side decreases with $c$ and the right side increases. At $c=a+|b-e|$, assume $b\\ge e$ by reversing the cycle; the required inequality reduces to $ae\\le b(a+b-e)$, which follows from\n$b(a+b-e)-ae=(b-e)(a+b)\\ge0$. Zero distances follow by continuity. This proves \\eqref{eq:snowflake}.\n\\end{proof}\nThis classical metric fact is included with proof; it is not claimed as new. See also \\cite{FS}.", "source_section": "2", "verification_method": "standalone_manuscript", "verification_status": "source_text", "artifact_paths": ["paper/Ilyoxune_Aggregation_Theorem.pdf", "paper/Ilyoxune_Aggregation_Theorem.tex"], "payload_json": "{\"content_format\": \"latex\", \"section_number\": \"2\"}"}
3
+ {"id": "section-3", "record_type": "manuscript_section", "title": "The sharp nonnegative congruence theorem", "content": "\\section{The sharp nonnegative congruence theorem}\\label{sec:quadratic}\nThis section controls every nonnegative linear substitution, including broadcasting a variable to several outputs and producing repeated monomials.\n\n\\begin{definition}\nLet $\\mathcal C_q$ be the set of symmetric entrywise nonnegative matrices $C$ for which\n\\[\n C_{ij}C_{k\\ell},\\quad C_{ik}C_{j\\ell},\\quad C_{i\\ell}C_{jk}\n\\]\nis $\\T_q$-null for every index quadruple, with repetitions allowed.\n\\end{definition}\n\n\\begin{theorem}[Sharp nonnegative congruence]\\label{thm:congruence}\nIf $C\\in\\mathcal C_q$ and $A$ is any entrywise nonnegative rectangular matrix, then\n\\[\n A^{\\mathsf T}CA\\in\\mathcal C_{\\max(q,1)}.\n\\]\nThe parameter $\\max\\{q,1\\}$ is optimal as a universal rule. In particular $\\mathcal C_q$ is preserved by every nonnegative congruence when $q\\ge1$.\n\\end{theorem}\n\\begin{proof}\nSet $B=A^{\\mathsf T}CA$, and fix any output quartet $a,b,c,d$, possibly repeated. For each ordered source quadruple define\n\\[\n W_{ijkl}=A_{ia}A_{jb}A_{kc}A_{\\ell d}.\n\\]\nThe three output products expand with exactly the same nonnegative weights:\n\\begin{align*}\n B_{ab}B_{cd}&=\\sum_{i,j,k,\\ell}W_{ijkl}C_{ij}C_{k\\ell},\\\\\n B_{ac}B_{bd}&=\\sum_{i,j,k,\\ell}W_{ijkl}C_{ik}C_{j\\ell},\\\\\n B_{ad}B_{bc}&=\\sum_{i,j,k,\\ell}W_{ijkl}C_{i\\ell}C_{jk}.\n\\end{align*}\nFor each source quadruple, take the $q$th roots of its three weighted terms. They satisfy all triangle inequalities. Lemma~\\ref{lem:minkowski} proves the three required inequalities at $r=\\max\\{q,1\\}$ after summing. No positivity, invertibility, or restriction on the shape of $A$ is used.\n\nTo force $p\\ge q$, take the absolute-minor $q$-power array of the real columns $(1,0),(0,1),(1,1),(1,-1)$ and choose the identity substitution. Its extremal quartet products are $(2^q,1,1)$.\nTo force $p\\ge1$, start with $F=x_1x_2$, valid at every $q$, and substitute\n\\[\n x_1=y_1+z,\\qquad x_2=y_2+z.\n\\]\nThe output is $y_1y_2+y_1z+y_2z+z^2$. Its normalized repeated-$z$ quartet products are exactly $(2,1,1)$, requiring $2^{1/p}\\le2$, or $p\\ge1$. Both lower bounds are attained by the stated upper bound.\n\\end{proof}\n\n\\begin{lemma}[Support is automatic]\\label{lem:quadraticsupport}\nThe nonzero quadratic coefficient support of a matrix in $\\mathcal C_q$ is M-convex.\n\\end{lemma}\n\\begin{proof}\nDelete zero rows. On the remaining indices define $i\\sim j$ when $i=j$ or $C_{ij}=0$. Suppose $C_{ij}=C_{jk}=0$ but $C_{ik}>0$. Since row $j$ is nonzero, choose $\\ell$ with $C_{j\\ell}>0$. The quartet $(i,k,j,\\ell)$ has exactly one positive product, a contradiction. Thus $\\sim$ is an equivalence relation. If $i\\ne j$ are in one class, choose a positive neighbor of $j$; the quartet $(i,i,j,\\ell)$ forces $C_{ii}=0$. Positive diagonal entries therefore occur only in singleton classes.\n\nThe support consists exactly of degree-two integer vectors having total at most one on every zero-diagonal class, and total at most two on each positive-diagonal singleton. These partition-capacity bases satisfy symmetric exchange. Indeed, for $x_i>y_i$, first choose a deficit coordinate in the same block if one exists. Otherwise $x$ has greater total than $y$ on that block, so choose a deficit coordinate in a block where $x$ has smaller total. The exchange respects both capacities for both vectors. Hence the support is M-convex.\n\\end{proof}\n\nFor a usual quadratic weak representation, the capacity-admissible quartet conditions imply the unrestricted matrix conditions defining $\\mathcal C_q$. If no coordinate is mandatory, an excluded repetition has zero diagonal at a capacity-one coordinate or a zero row at a capacity-zero coordinate; its list is zero, an equal pair, or an equal triple. If a coordinate is mandatory, the degree-two support is a star or one monomial, and every unrestricted quartet is again one of these tautological types. Thus the following corollary covers all quadratic weak representations, rather than just a restricted matrix subclass.\n\n\\begin{corollary}[Exact quadratic substitution calculus]\nWrite $F(x)=\\tfrac12x^{\\mathsf T}Cx$ with normalized quadratic coefficient matrix $C\\in\\mathcal C_q$. For every nonnegative linear substitution $x=Ay$, the full normalized output matrix is exactly $A^{\\mathsf T}CA$ and if nonzero, its weak and strong parameter is universally repaired by $p=\\max\\{q,1\\}$. No parameter loss accumulates under any finite composition once the parameter is at least one.\n\\end{corollary}\n\\begin{proof}\nThe matrix identity follows by substitution. The support lemma supplies M-convexity. All rank-two strong relations have at most three terms, so the quartet conditions give both weak and strong representations. A composition of nonnegative matrices is nonnegative; apply Theorem~\\ref{thm:congruence} to their product.\n\\end{proof}", "source_section": "3", "verification_method": "standalone_manuscript", "verification_status": "source_text", "artifact_paths": ["paper/Ilyoxune_Aggregation_Theorem.pdf", "paper/Ilyoxune_Aggregation_Theorem.tex"], "payload_json": "{\"content_format\": \"latex\", \"section_number\": \"3\"}"}
4
+ {"id": "section-4", "record_type": "manuscript_section", "title": "Uniform obstructions in every rank and size", "content": "\\section{Uniform obstructions in every rank and size}\\label{sec:uniform}\nChoose six distinguished labels $1,\\ldots,6$ and a disjoint set $C$ of size $d-3$. The ground set may contain any additional labels. Put\n\\[\n U=C\\cup135,\\qquad Z_1=C\\cup126,\\qquad Z_2=C\\cup346.\n\\]\n\n\\begin{lemma}[Three exceptional bases]\\label{lem:uniform}\nFor $1<b<2$ and $0<e\\le1/2$, set $a_U=b$, $a_{Z_1}=a_{Z_2}=e$, and $a_S=1$ on every other $d$-subset of an $n$-element ground set, $n\\ge d+3$. For every $q>0$, the coefficient vector $c_S=a_S^q$ is strict strong over $\\T_q$ on $U_{d,n}$. Every genuine root-polygon slack is at least\n\\begin{equation}\\label{eq:slack}\n \\delta=\\min\\{e^2,2-b\\}.\n\\end{equation}\n\\end{lemma}\n\\begin{proof}\nAfter taking roots, each strong term is a product of two $a$'s. In a relation with $m\\ge3$ terms, each exceptional coefficient can occur at most once among the factors: two different first factors are distinct, two different second factors are distinct, and a first factor cannot equal a second factor when $m\\ge3$. Every term is at most two; a term containing a low coefficient is at most $be<1$.\n\nFor $m=3$, factors from different terms intersect in $d-1$ elements. The three exceptional bases intersect pairwise in $d-2$ elements. Consequently any two exceptional factors occurring in the relation must occur in the same term. At least two terms therefore equal one.\n\nFor $m=4$, if fewer than three terms are exceptional there are again two unit terms. If all three exceptional bases occur, the union $I\\cup J$ has size $d+3$ and equals $C\\cup[6]$. Complement each basis inside this union. The dual relation has rank three and its exceptional triples are a peak at $246$ and lows at $125,345$. The permutation $(12)(34)(56)$ takes them to the original peak at $135$ and lows at $126,346$. The rank-three four-term products have at least two unit terms. To see this directly, complementary products at the sparse limit have possible lists\n\\[\n (1,1,1,1),\\ (0,1,1,1),\\ (0,0,1,1),\\ (1,1,1,2),\\ (0,1,1,2).\n\\]\nThe two zeros can occur together only when $I=12$ or $34$, in which case the peak product cannot occur. Restoring positive lows and replacing the peak root by $b$ retains the two unit terms.\n\nFor $m\\ge5$, there are at most three exceptional terms, so at least $m-3\\ge2$ unit terms remain. Thus in every case a peak term has slack at least $2-b$. Without a peak term the largest term is at most one, two unit terms remain, and at least one additional term is positive and at least $e^2$. This proves \\eqref{eq:slack} and strictness.\n\\end{proof}\n\n\\begin{theorem}[Explicit all-rank obstruction]\\label{thm:explicit}\nFor every $q>0$, define\n\\begin{equation}\\label{eq:parameters}\n T=2^q,\\quad D=2^{-q}-4^{-q},\\quad\n s=\\frac{qD}{64(q+2)},\\quad\n L=\\frac{3D^2}{128(q+2)},\\quad b=2-\\frac{L}{16}.\n\\end{equation}\nUse the root array of Lemma~\\ref{lem:uniform} with $e=1/4$ and put\n\\begin{equation}\\label{eq:explicit}\n c_S=a_S^q s^{\\one_{6\\in S}}.\n\\end{equation}\nFor every $d\\ge3,n\\ge d+3$, this is positive strict strong on $U_{d,n}$, whereas merging $5,6$ fails a squarefree weak relation by at least $L/2$ in root lengths.\n\\end{theorem}\n\\begin{proof}\nSince $0<D\\le1/4$, we have $L<1$, $1<b<2$, and $0<s\\le\\min(1,q)$. Lemmas~\\ref{lem:uniform} and \\ref{lem:basic} prove the input assertion.\n\nIn the output use common index $C\\cup\\{z\\}$ and quartet $1,2,3,4$. Omitting the common $C$ from notation, the six coefficients are\n\\begin{equation}\\label{eq:six}\n g_{12}=g_{34}=1+4^{-q}s,\\qquad\n g_{14}=g_{23}=g_{24}=1+s,\\qquad g_{13}=b^q+s.\n\\end{equation}\nTemporarily replace $b$ by $2$ and define\n\\[\n H(t)=((T+t)(1+t))^{1/q}-(1+t)^{2/q}-(1+4^{-q}t)^{2/q}.\n\\]\nThen $H(0)=0$ and $H'(0)=2D/q$. Put $r_0=1/q$. For $0\\le t\\le\\min(1,q)$, the three positive expressions defining $H$ are at most $2\\exp(2),\\exp(2),\\exp(2)$. Logarithmic differentiation bounds the absolute second derivative of each expression by its value times $4r_0^2+2r_0$. Since $\\exp(2)<8$,\n\\[\n |H''(t)|<64r_0(2r_0+1)=M.\n\\]\nThe selected $s$ satisfies $Ms=r_0D$. Taylor's integral remainder gives\n\\[\n H(s)\\ge2r_0Ds-\\tfrac12Ms^2\n \\ge\\tfrac32r_0Ds=L.\n\\]\nOnly the first expression changes when the peak root decreases from $2$ to $b$. For $1\\le v\\le2$ its derivative is\n\\[\n \\frac{\\partial}{\\partial v}((v^q+s)(1+s))^{1/q}\n =(1+s)^{r_0}(1+s/v^q)^{r_0-1}<8.\n\\]\nIndeed, if $r_0-1<0$ its second factor is at most one; otherwise the product is at most $\\exp(2r_0s)\\le\\exp(2)<8$. Decreasing $v$ by $L/16$ loses at most $L/2$. Thus\n\\[\n (g_{13}g_{24})^{1/q}-(g_{14}g_{23})^{1/q}-(g_{12}g_{34})^{1/q}\\ge L/2>0.\n\\]\nExtra labels never enter these six coefficients, proving the claim in every $d,n$.\n\\end{proof}\n\n\\begin{corollary}[Explicit robustness independent of dimension]\\label{cor:robust}\nLet $c$ be \\eqref{eq:explicit} and put $\\eta=L/2048$. Every positive array $c'$ satisfying\n\\[\n 1-\\eta\\le(c'_S/c_S)^{1/q}\\le1+\\eta\\qquad\\text{for every }S\n\\]\nis still strict strong and still fails after the same merge. Its normalized input root-polygon slack is at least $3L/256$, and its output defect is at least $3L/32$.\n\\end{corollary}\n\\begin{proof}\nFactor the common diagonal-rescaling multiplier from each input relation. Each root product changes by a factor in $[(1-\\eta)^2,(1+\\eta)^2]$. Its original maximum is at most two and its slack at least $L/16$. For any chosen side the new slack is at least\n\\[\n (1-\\eta)^2\\frac L{16}-8\\eta\n \\ge\\frac L{64}-\\frac L{256}=\\frac{3L}{256}.\n\\]\nThe same relative bounds hold on the roots of merged coefficient products: sums preserve the coefficient bounds before taking roots. The sum of the two smaller original output root lengths is less than $16$, using the bounds in the preceding proof. The new defect is therefore at least\n\\[\n (1-\\eta)^2\\frac L2-4\\eta\\cdot16\n \\ge\\frac L8-\\frac L{32}=\\frac{3L}{32}.\n\\]\n\\end{proof}", "source_section": "4", "verification_method": "standalone_manuscript", "verification_status": "source_text", "artifact_paths": ["paper/Ilyoxune_Aggregation_Theorem.pdf", "paper/Ilyoxune_Aggregation_Theorem.tex"], "payload_json": "{\"content_format\": \"latex\", \"section_number\": \"4\"}"}
5
+ {"id": "section-5", "record_type": "manuscript_section", "title": "A guaranteed repair for the full normalized output", "content": "\\section{A guaranteed repair for the full normalized output}\\label{sec:repair}\nThe negative result concerns keeping $q$ unchanged. The next theorem shows that controlled weakening of the output condition always repairs \\emph{weak} compatibility for positive uniform input.\n\n\\begin{theorem}[One-merge repair]\\label{thm:repair}\nLet $c_S>0$ for every $d$-subset of $[n]$, and suppose $c$ is weak over $\\T_q$. For $r=\\max\\{q,1\\}$, the normalized coefficient function \\eqref{eq:rho} of $D_{u,v}F$ is weak over $\\T_{2r}$. If $d=2$, $\\T_r$ suffices. Every larger output parameter also suffices.\n\\end{theorem}\n\\begin{proof}\nThe support is the capped simplex described in Section~\\ref{sec:definitions}, so it remains to verify all tests \\eqref{eq:generalweak}. Let $s_0=\\alpha_z$ and let $k_0$ be the multiplicity of $z$ in the quartet. Capacities give $s_0+k_0\\le2$, so there are exactly six cases.\n\n\\textbf{Case $(s_0,k_0)=(0,0)$.} The products are an original input relation. They remain null at $2r\\ge q$ by monotonicity.\n\n\\textbf{Case $(0,1)$.} Write $K$ for the squarefree common index. Each output product is a sum of two products, one using $u$, one using $v$. For example,\n\\[\n c_{Kij}(c_{Kku}+c_{Kkv})=\\sum_{h\\in\\{u,v\\}}c_{Kij}c_{Kkh}.\n\\]\nFor each fixed $h$, the three $q$th roots form a triangle by the input relation on $i,j,k,h$. Lemma~\\ref{lem:minkowski} gives the output triangle at $p=r$, and hence at $2r$.\n\n\\textbf{Case $(0,2)$.} Let $K$ be the common index and $i,j$ the two nonmerged quartet labels. Put $a=c_{Kij}$, $w=c_{Kuv}$, and $b_i=c_{Kiu}+c_{Kiv}$, with $b_j$ defined similarly. The input quartet $i,j,u,v$ gives\n\\[\n (aw)^{1/q}\\le(c_{Kiu}c_{Kjv})^{1/q}+(c_{Kiv}c_{Kju})^{1/q}.\n\\]\nUsing $(x+y)^q\\le2^{r-1}(x^q+y^q)$ for $q\\ge1$, and subadditivity for $q\\le1$, yields\n\\[\n 2aw\\le2^r(c_{Kiu}c_{Kjv}+c_{Kiv}c_{Kju})\\le2^r b_i b_j.\n\\]\nThe output products are $2aw,b_i b_j,b_i b_j$. Thus they form a null list already at $p=r$.\n\n\\textbf{Case $(1,0)$: the only parameter-doubling case.}\nWrite $\\alpha=C+z$, $|C|=d-3$, and let the quartet be $i,j,k,\\ell$. Define\n\\[\n a^h_{ij}=c_{Cijh}^{1/q}\\quad(h=u,v),\\qquad w_i=c_{Ciuv}^{1/q}>0.\n\\]\nApply the input weak relation with common index $C\\cup\\{u\\}$ and quartet $i,j,k,v$. One of its triangle inequalities is\n\\[\n a^u_{ij}w_k\\le a^u_{ik}w_j+a^u_{jk}w_i.\n\\]\nThe analogous statement holds for $v$. Hence\n\\[\n d^h_{ij}=\\frac{a^h_{ij}}{w_iw_j},\\qquad d^h_{ii}=0,\n\\]\nare metrics on the unmerged labels under consideration. Positivity is essential here and is part of the theorem's hypotheses. For $q\\ge1$ their pointwise $\\ell_q$ aggregate\n\\[\n D_{ij}=((d^u_{ij})^q+(d^v_{ij})^q)^{1/q}\n\\]\nis a metric by Minkowski. For $q\\le1$, each $(d^h)^q$ is a metric, so $D_{ij}=(d^u_{ij})^q+(d^v_{ij})^q$ is a metric. In both cases the merged coefficients satisfy\n\\begin{equation}\\label{eq:metricfactor}\n b_{ij}:=c_{Ciju}+c_{Cijv}=(w_iw_j)^qD_{ij}^{r}.\n\\end{equation}\nAt parameter $p=2r$, the root product associated with a pairing of the quartet equals\n\\[\n (b_{ij}b_{k\\ell})^{1/(2r)}\n =(w_iw_jw_kw_\\ell)^{q/(2r)}\\sqrt{D_{ij}D_{k\\ell}}.\n\\]\nThe common positive multiplier cancels. Lemma~\\ref{lem:snowflake} supplies precisely all three required inequalities.\n\n\\textbf{Case $(1,1)$.} Write $\\alpha=C+z$ and use quartet $i,j,k,z$. Set $W_i=c_{Ciuv}$ and $B_{jk}=c_{Cjku}+c_{Cjkv}$. The three output products are\n\\[\n 2W_iB_{jk},\\quad2W_jB_{ik},\\quad2W_kB_{ij}.\n\\]\nFor each fixed $h=u,v$, the products $W_i c_{Cjkh}$, $W_j c_{Cikh}$, $W_k c_{Cijh}$ form an input weak relation with common index $C\\cup\\{h\\}$. Aggregate the two lists using Lemma~\\ref{lem:minkowski} and discard the shared factor two. This proves nullity at $p=r$.\n\n\\textbf{Case $(2,0)$.} The common index has the form $C+2z$. Each coefficient is twice an original coefficient with $u,v$ present. All three products are multiplied by four, and the input weak relation with common index $C\\cup\\{u,v\\}$ applies.\n\nThese six cases exhaust all admissible tests. In degree two, $|\\alpha|=0$, so only the first three cases occur and $p=r$ suffices throughout. Monotonicity proves the assertion for larger $p$.\n\\end{proof}\n\n\\begin{theorem}[Arbitrary matroid support]\\label{thm:zeros}\nThe normalized output of one merge of any weak $\\T_q$ matroid coefficient array is weak over $\\T_{2\\max\\{q,1\\}}$. In degree two, $\\T_{\\max\\{q,1\\}}$ suffices, with strong output as well.\n\\end{theorem}\n\\begin{proof}\nThe output support is M-convex by Lemma~\\ref{lem:support} below. The five elementary cases in the preceding proof allow zeros unchanged. Consider its critical case $(1,0)$ and fix its quartet. Retain the notation $a^h_{ij}=c_{Cijh}^{1/q}$ and $w_i=c_{Ciuv}^{1/q}$.\n\nIf every $w_i>0$, the preceding metric argument applies directly with pseudometrics: its divisions are valid, and vanishing distances do not affect Minkowski or Lemma~\\ref{lem:snowflake}. Hence assume that at least one weight vanishes.\n\nSuppose some $w_i$ is positive. The weighted triangle inequalities imply that an edge between two zero-weight vertices has $a^h_{ij}=0$: use a positive third vertex. They also imply that for a zero-weight vertex $i$ and positive-weight vertices $j,k$, $a^h_{ij}/w_j=a^h_{ik}/w_k=:A_{i,h}$, by applying both directions of the same inequality. If at least two weights vanish, the output products are either all zero or $(0,B,B)$, so are null at every parameter.\n\nIf exactly one weight, say $w_i$, vanishes, put $M=A_{i,u}^q+A_{i,v}^q$. For positive vertices $j$ we have $b_{ij}=w_j^q M$. On the other three vertices the aggregate metric $D$ in \\eqref{eq:metricfactor} is defined as before, allowing zero distances. If $M=0$ all output products vanish. Otherwise each output root product at parameter $2r$ is a common positive multiple of one of $\\sqrt{D_{jk}},\\sqrt{D_{j\\ell}},\\sqrt{D_{k\\ell}}$. These form a triangle, since square roots of metric distances satisfy the triangle inequality. Lemma~\\ref{lem:snowflake} also holds for pseudometrics by the same proof.\n\nIt remains to treat $w_i=0$ at all four vertices. The input weak relation with common index $C\\cup\\{i\\}$ and quartet $j,k,u,v$ now has two nonzero candidate products, and forces\n\\[\n c_{Ciju}c_{Cikv}=c_{Cijv}c_{Ciku}.\n\\]\nThus nonzero coefficient two-vectors on adjacent edges are proportional. Suppose two disjoint nonzero edge vectors $v_{ij}=(c_{Ciju},c_{Cijv})$ and $v_{k\\ell}$ were not proportional. Every bridge vector $v_{ik},v_{i\\ell},v_{jk},v_{j\\ell}$ is adjacent to both and must be zero. Choose nonzero coordinates $h,g\\in\\{u,v\\}$ to obtain actual matroid bases $C\\cup\\{i,j,h\\}$ and $C\\cup\\{k,\\ell,g\\}$. Basis exchange removing $i$ from the first basis is impossible: exchanging with $k$ or $\\ell$ gives a zero bridge coefficient, and exchanging with $g\\ne h$ gives $c_{Cjuv}=0$. This contradicts the support matroid's basis-exchange axiom.\n\nAll six nonnegative edge vectors therefore lie on one ray (zero vectors allowed). If every vector is zero there is nothing to prove. Otherwise choose a component $h$ positive on that ray. There is a single scalar $\\lambda>0$ such that $b_{ij}=\\lambda c_{Cijh}$ for every edge. The output relation is a rescaled input weak relation, valid already at $q$ and hence at $2r$. This completes all zero cases without a density assumption. Degree two never uses this critical case, so its earlier bound holds for every rank-two matroid.\n\\end{proof}\n\n\\begin{remark}[Scope of the repair]\nThe output is weak; strong output in degree at least three is not proved. The arbitrary-support extension uses the matroid support axiom in its final zero case; no positivity or strong-input hypothesis is needed. Weighted two-variable merges are covered by absorbing their positive weights into diagonal input rescalings.\n\\end{remark}", "source_section": "5", "verification_method": "standalone_manuscript", "verification_status": "source_text", "artifact_paths": ["paper/Ilyoxune_Aggregation_Theorem.pdf", "paper/Ilyoxune_Aggregation_Theorem.tex"], "payload_json": "{\"content_format\": \"latex\", \"section_number\": \"5\"}"}
6
+ {"id": "section-6", "record_type": "manuscript_section", "title": "Sharp thresholds and asymptotically optimal parameter cost", "content": "\\section{Sharp thresholds and asymptotically optimal parameter cost}\\label{sec:optimal}\n\\subsection{Determinants as certified strong inputs}\n\\begin{lemma}\\label{lem:det}\nFor a real rank-$d$ matrix $A$ and any $q>0$, $c_S=|\\det A_S|^q$ is strong over $\\T_q$ on its column matroid.\n\\end{lemma}\n\\begin{proof}\nThe signed determinant Pl\\\"ucker relation sums to zero. The absolute value of each term is at most the sum of the others by the ordinary triangle inequality. These absolute values are exactly the $q$th roots of the products in \\eqref{eq:strong}.\n\\end{proof}\n\nTake four planar points $u_i\\in\\R^2$, their affine lifts $(u_i,1)$, and two additional columns $e_1,e_2\\in\\R^3$. After merging the last two labels,\n\\begin{equation}\\label{eq:planar}\n b_{ij}=|u_{i,1}-u_{j,1}|^q+|u_{i,2}-u_{j,2}|^q.\n\\end{equation}\nThis directly verifies the required output coefficients; no metric theorem is needed for the lower bounds below.\n\n\\begin{proposition}[Necessary output parameter]\\label{prop:lower}\nAny universal parameter $p$ for one merge in degree at least three, even restricted to positive uniform strong $\\T_q$ inputs, must satisfy\n\\begin{equation}\\label{eq:lower}\n p\\ge\\max\\{2,2(q-1)\\}.\n\\end{equation}\nMoreover, in each fixed degree $d\\ge3$ and size $n\\ge d+3$, the infimal universal parameter is strictly greater than $q$.\n\\end{proposition}\n\\begin{proof}\nFor the planar unit square, \\eqref{eq:planar} gives adjacent coefficients one and diagonal coefficients two, independently of $q$. The three output products are $(1,4,1)$. The output $\\T_p$ relation holds exactly when $4^{1/p}\\le2$, or $p\\ge2$.\n\nFor the coordinate diamond $(1,0),(0,1),(-1,0),(0,-1)$ the three products are $(4,4^q,4)$. When $q>1$, the necessary inequality $4^{q/p}\\le2\\cdot4^{1/p}$ is equivalent to $p\\ge2(q-1)$; for $q\\le1$ this bound is vacuous.\n\nThese matrices may have zero minors. Whenever a proposed bound is violated, the failed output inequality is strict. A sufficiently small generic matrix perturbation makes every maximal minor nonzero and preserves this strict failure, while Lemma~\\ref{lem:det} certifies strong input at the fixed $q$. To realize every $d,n$, first add $d-3$ identity columns and rows, then any extra columns, and perturb generically. The selected quartet minors are nonzero before perturbation, so the same failure persists.\n\nFinally, Theorem~\\ref{thm:explicit} provides, at each fixed $q,d,n$, a witness with a positive output defect at parameter $q$. That witness continues to fail for all output parameters in some interval $[q,q+\\eps]$ by continuity. Thus the infimum of universal working parameters is at least $q+\\eps>q$.\n\\end{proof}\n\n\\subsection{Degree two is completely determined}\n\\begin{theorem}[Exact degree-two answer]\\label{thm:d2}\nFor degree-two matroid input, the infimal universal output parameter for one merge is\n\\[\n P_2(q)=\\max\\{q,1\\}.\n\\]\nThe output is both weak and strong at this parameter. Consequently the same-parameter threshold in degree two is exactly $q_2=1$.\n\\end{theorem}\n\\begin{proof}\nThe upper bound is Theorems~\\ref{thm:repair} and \\ref{thm:zeros}. In rank two every strong relation has at most three summands, so weak and strong coincide. Equivalently, polarize the output by replacing the capacity-two coordinate $z$ with two clones. Its coefficient on the two clones is $\\rho_{2z}$ and its coefficients on a clone and $i$ are $\\rho_{zi}$. Every rank-two strong test is a three-term quartet, including the repeated-$z$ test already checked.\n\nFor $p\\ge q$ necessity use four real columns $(1,0),(0,1),(1,1),(1,-1)$: their Pl\\\"ucker product magnitudes are $(2,1,1)$. After raising minors to $q$, the products are $(2^q,1,1)$, which require $p\\ge q$. Add two very small generic columns and merge those columns; the untouched quartet is unchanged and all input minors can be nonzero.\n\nFor $p\\ge1$ necessity take the five real columns\n\\[\n e_1,\\quad e_2,\\quad e_1+e_2,\\quad e_1,\\quad e_2\n\\]\nand merge the last two. On the unmerged labels $1,2,3$ the coefficients are all one. The merged coefficients are $b_1=b_2=1,b_3=2$, for every input exponent $q>0$. The output quartet products are $(2,1,1)$ and require $p\\ge1$. A small generic perturbation gives positive uniform input while retaining strict failure for any fixed proposed $p<1$. Extra columns can be added if desired. The two lower bounds and the upper bound agree.\n\\end{proof}\nDegree one has only linear coefficients; identification preserves it at every parameter.\n\n\\subsection{Exact low-parameter repair and the factor two}\nFor fixed $d\\ge3,n\\ge d+3$, let $P^{\\rm str}_{d,n}(q)$ be the infimum of parameters $p$ for which every positive uniform strong $\\T_q$ input has normalized weak $\\T_p$ output after one specified merge. Define $P^{\\rm w}_{d,n}(q)$ similarly for weak input. Monotonicity makes the universal working sets upper intervals, and Theorem~\\ref{thm:repair} makes them nonempty.\n\n\\begin{corollary}[Two-sided aggregation law]\\label{cor:cost}\nFor $d\\ge3,n\\ge d+3$,\n\\[\n P^{\\rm str}_{d,n}(q)\\le P^{\\rm w}_{d,n}(q)\\le2\\max\\{q,1\\},\n\\]\n\\[\n P^{\\rm str}_{d,n}(q)\\ge\\max\\{2,2(q-1)\\},\n \\qquad P^{\\rm str}_{d,n}(q)>q.\n\\]\nIn particular,\n\\[\n P^{\\rm str}_{d,n}(q)=P^{\\rm w}_{d,n}(q)=2\\quad(0<q\\le1),\n\\]\nand\n\\[\n \\lim_{q\\to\\infty}\\frac{P^{\\rm str}_{d,n}(q)}q\n =\\lim_{q\\to\\infty}\\frac{P^{\\rm w}_{d,n}(q)}q=2.\n\\]\nFor $q\\ge1$, the additive gap between the stated upper and lower bounds is at most two.\n\\end{corollary}\n\\begin{proof}\nCombine Theorem~\\ref{thm:repair} and Proposition~\\ref{prop:lower}. For $q\\ge2$, the bounds $2q-2$ and $2q$ squeeze the ratios to two and have difference two. For $1\\le q\\le2$, the lower bound two and upper bound $2q$ also have difference at most two.\n\\end{proof}\nThe full higher-degree optimum remains undetermined by the elementary arguments here. Section~\\ref{sec:peaks} strengthens the lower bounds and gives a quantitative obstruction at $q=2$.", "source_section": "6", "verification_method": "standalone_manuscript", "verification_status": "source_text", "artifact_paths": ["paper/Ilyoxune_Aggregation_Theorem.pdf", "paper/Ilyoxune_Aggregation_Theorem.tex"], "payload_json": "{\"content_format\": \"latex\", \"section_number\": \"6\"}"}
7
+ {"id": "section-7", "record_type": "manuscript_section", "title": "The exact minimal obstruction boundary", "content": "\\section{The exact minimal obstruction boundary}\\label{sec:minimal}\nThe universal failure in rank at least three first appears at corank three. Below it, the full normalized output retains strong compatibility in a precisely determined parameter range.\n\nFor clarity, strong compatibility of the one-merge output can be tested by its natural polarization: replace the capacity-two variable $z$ by two labeled clones, and assign to a $d$-subset the normalized coefficient $\\rho_\\beta$ with the same count vector $\\beta$. The clone array has $n$ ground labels. Weak clone relations are exactly the bounded-multiplicity quartet tests of the output, with the usual zero or equal-pair tautologies. Its support is a matroid. For a direct proof, take two clone bases with count vectors $x,y$ and a clone in the first but not the second, in block $h$. If $x_h\\le y_h$, swap it with an unused clone of the second basis in the same block. If $x_h>y_h$, M-convex exchange supplies a deficit block $j$ and valid exchanged counts; choose a clone in the second but not the first in that block. This verifies ordinary basis exchange for the lift. Lemma~\\ref{lem:support} proves the needed M-convexity.\n\n\\begin{theorem}[Complete one-merge preservation classification]\\label{thm:minimal}\nFix $q>0$, rank $d\\ge1$, and $n\\ge\\max(d,2)$, with a specified pair to be merged. Universal same-parameter preservation from strong input to strong full normalized output holds exactly in the regimes of Table~\\ref{tab:threshold}. Replacing ``strong output'' by ``weak output'' gives the identical classification. Every excluded regime admits a uniform positive input counterexample.\n\\end{theorem}\n\\begin{proof}\nDegree one is immediate. For $d\\ge2$ and $n\\le d+1$, the polarized output has corank at most one. A strong relation has\n\\[\n |J\\setminus I|\\le n-d+1\\le2\n\\]\nterms, hence is zero or a tautological equal pair. Thus every parameter works.\n\nSuppose $n=d+2$. The critical metric case $(\\alpha_z,k_z)=(1,0)$ in Theorem~\\ref{thm:repair} requires a common set of $d-3$ labels, four quartet labels, and the two merged labels: $d+3$ distinct source elements. It is impossible here. All five remaining cases work already at $p=\\max\\{q,1\\}$, including their zero cases. For $q\\ge1$ the parameter stays $q$. The polarized output now has corank two, so all its strong relations have at most three terms and weak equals strong.\n\nFor $d=2$ and any $n\\ge4$, Theorem~\\ref{thm:congruence} gives same-parameter strong preservation whenever $q\\ge1$.\n\nFor the converse at $q<1$, choose $M>0$ and the rank-two real columns\n\\[\n i=(1,0),\\quad j=(0,1),\\quad\n u=(1,M^{1/q}),\\quad v=(M^{1/q},-1).\n\\]\nTheir absolute-minor $q$-power coefficients are strong by Lemma~\\ref{lem:det}. Merging $u,v$ gives the normalized output product list\n\\[\n 2(1+M^{2/q})^q,\\qquad (M+1)^2,\\qquad(M+1)^2.\n\\]\nThe ratio of the first product to the other two tends to $2$ as $M\\to\\infty$. Nullity at parameter $q$ would require that ratio to be at most $2^q<2$, so sufficiently large $M$ gives strict failure. All six source minors are nonzero. Adding $d-2$ identity coloops gives rank $d$ on $d+2$ elements; a sufficiently small generic real matrix perturbation makes every maximal minor nonzero while preserving the strict failed relation. For degree two, additional generic columns cover every $n>4$ as well.\n\nFinally, if $d\\ge3,n\\ge d+3$, Theorem~\\ref{thm:explicit} supplies a strict positive uniform counterexample at every $q>0$. This exhausts the table and proves both classifications.\n\\end{proof}\n\n\\begin{example}[Smallest exact failure below one]\nAt $q=1/2$ and $M=8$, the four real columns above are\n$(1,0),(0,1),(1,64),(64,-1)$. The nontrivial input root products are $(4097,4096,1)$, so the input is strong. The merged output root products at $q=1/2$ are\n\\[\n 16388,\\qquad6561,\\qquad6561.\n\\]\nThe excess is the integer $3266>0$. Thus failure already occurs on $U_{2,4}$, the smallest possible nontrivial ground set in degree two.\n\\end{example}", "source_section": "7", "verification_method": "standalone_manuscript", "verification_status": "source_text", "artifact_paths": ["paper/Ilyoxune_Aggregation_Theorem.pdf", "paper/Ilyoxune_Aggregation_Theorem.tex"], "payload_json": "{\"content_format\": \"latex\", \"section_number\": \"7\"}"}
8
+ {"id": "section-8", "record_type": "manuscript_section", "title": "Two peaks: exact weak versus strong compatibility", "content": "\\section{Two peaks: exact weak versus strong compatibility}\\label{sec:peaks}\nFix $d\\ge3,n\\ge d+3$ and $|C|=d-3$ disjoint from $[6]$. Set $U=C\\cup135$ and $V=C\\cup246$. These two exceptional bases meet in $d-3$ elements.\n\n\\begin{theorem}[Exact two-peak classification]\\label{thm:twopeak}\nLet $a,b\\ge1$, define root coefficients $a_U=a$, $a_V=b$, and $a_S=1$ otherwise, and put $c_S=a_S^q$. Then\n\\[\n c\\text{ is weak over }\\T_q\\quad\\Longleftrightarrow\\quad a\\le2,\\ b\\le2,\n\\]\n\\[\n c\\text{ is strong over }\\T_q\\quad\\Longleftrightarrow\\quad\n a\\le2,\\ b\\le2,\\ ab\\le3.\n\\]\nAll genuine weak inequalities are strict when $a,b<2$, and all genuine strong inequalities are strict when additionally $ab<3$.\n\\end{theorem}\n\\begin{proof}\nA weak relation has union size $d+2$, whereas $U\\cup V$ has size $d+3$. Thus it contains at most one peak, and its root products are $(1,1,1)$, $(a,1,1)$, or $(b,1,1)$. Both peak types occur; this proves the weak equivalence and its strictness statement.\n\nFor a four-term strong relation, exceptional factors cannot lie in different terms: factors in different terms intersect in either $d-1$ or $d-2$ elements, whereas $|U\\cap V|=d-3$. If both peaks occur they pair together and give $(ab,1,1,1)$. Such a relation exists, with $I=C\\cup\\{1,3\\}$ and $J=C\\cup\\{2,4,5,6\\}$. It forces $ab\\le3$. Relations with a single peak are harmless under $a,b\\le2$. A relation with at least five terms has at most two exceptional terms, retaining at least three unit terms; its largest term is at most two. Both peaks cannot occupy one product in this case, since the intersection of the two factors of a product would have to have size $d+1-m<d-3$. Therefore all these relations are strict. This proves sufficiency and necessity.\n\\end{proof}\n\nAfter merging $5,6$, four relevant coefficients equal two and the other two equal $1+a^q$ and $1+b^q$. Their product list is\n\\begin{equation}\\label{eq:peakoutput}\n 4,\\qquad(1+a^q)(1+b^q),\\qquad4.\n\\end{equation}\n\n\\begin{corollary}[Sharper weak repair lower bound]\\label{cor:weaklower}\nFor every $d\\ge3,n\\ge d+3$,\n\\[\n P^{\\rm w}_{d,n}(q)\\ge\n \\max\\left\\{2,\\;2\\log_2\\frac{1+2^q}{2}\\right\\}.\n\\]\nFor strong inputs one also has\n\\[\n P^{\\rm str}_{d,n}(q)\\ge\n \\log_2\\frac{(1+2^q)(1+(3/2)^q)}4.\n\\]\n\\end{corollary}\n\\begin{proof}\nTake $a=b=2$ for the weak assertion, and $a=2,b=3/2$ for the strong assertion, and solve the null inequality for \\eqref{eq:peakoutput}. The universal lower bound two is Proposition~\\ref{prop:lower}. Lowering the peaks slightly makes the required input inequalities strict while preserving every strict output failure, so the bounds also hold when testing only strict inputs.\n\\end{proof}\nThe weak bound exceeds $2(q-1)$ strictly for every $q>0$. It forces $P^{\\rm w}_{d,n}(q)>2$ whenever $q>\\log_2 3$. No matching upper curve is asserted here.\n\n\\begin{example}[Three integer levels at parameter two]\nOn every $U_{d,n}$ in this section, set $c_U=23$, $c_V=14$, and every other coefficient to $6$. After removing the common root scale $\\sqrt6$, the peaks are $\\sqrt{23/6}<2$ and $\\sqrt{14/6}<2$, with product $\\sqrt{322}/6<3$. The input is strict strong by Theorem~\\ref{thm:twopeak}. Its merged product list is $(580,144,144)$, which fails $\\T_2$ because\n\\[\n 580-(12+12)^2=4>0.\n\\]\n\\end{example}\n\n\\begin{theorem}[A quantitative strong-input barrier at two]\\label{thm:twentyninths}\nFor every $d\\ge3,n\\ge d+3$,\n\\[\n P^{\\rm str}_{d,n}(2)>\\frac{20}{9}.\n\\]\n\\end{theorem}\n\\begin{proof}\nUse the three exceptional bases $U=C\\cup135$, $Z_1=C\\cup126$, $Z_2=C\\cup346$. Assign\n\\[\n c_U=3999,\\quad c_{Z_1}=c_{Z_2}=1,\\quad\n c_S=5000\\ (6\\in S\\text{ otherwise}),\\quad\n c_S=1000\\ (6\\notin S\\text{ otherwise}).\n\\]\nThis is Lemma~\\ref{lem:uniform} at $q=2$, root peak $\\sqrt{3999/1000}<2$, root low $1/\\sqrt{5000}<1/2$, followed by scaling label six by five and globally by $1000$. Hence all genuine strong input inequalities are strict.\n\nThe merged coefficients are $g_{12}=g_{34}=1001$, $g_{14}=g_{23}=g_{24}=6000$, and $g_{13}=8999$. Divide their three products by $10^6$ to obtain\n\\[\n A=\\frac{26997}{500},\\qquad B=36,\\qquad D=\\frac{1002001}{1000000}.\n\\]\nExact integer-power comparisons give\n\\[\n A^9>\\left(\\frac{6019}{1000}\\right)^{20},\\quad\n 36^9<\\left(\\frac{5016}{1000}\\right)^{20},\\quad\n D^9<\\left(\\frac{1001}{1000}\\right)^{20}.\n\\]\nFor a short manual or executable check, multiply each inequality by its positive denominators. For example, the first is exactly\n$26997^9\\,1000^{20}>6019^{20}\\,500^9$; the other two are equally direct integer comparisons. They imply\n\\[\n A^{9/20}-B^{9/20}-D^{9/20}>\\frac{2}{1000}>0.\n\\]\nThus the fixed strict input fails at output parameter $20/9$. Continuity in the output parameter gives failure on a nonempty interval above $20/9$, proving the strict inequality for the infimal universal repair parameter. All selected relations are unchanged by extra ground labels.\n\\end{proof}", "source_section": "8", "verification_method": "standalone_manuscript", "verification_status": "source_text", "artifact_paths": ["paper/Ilyoxune_Aggregation_Theorem.pdf", "paper/Ilyoxune_Aggregation_Theorem.tex"], "payload_json": "{\"content_format\": \"latex\", \"section_number\": \"8\"}"}
9
+ {"id": "section-9", "record_type": "manuscript_section", "title": "Exact classification of weighted squarefree aggregation", "content": "\\section{Exact classification of weighted squarefree aggregation}\\label{sec:classification}\nFix $d\\ge3$ and $q>0$. Let an operator substitute\n\\[\n x_i\\longmapsto w_i y_{\\phi(i)},\\qquad w_i\\ge0,\n\\]\nand then take the squarefree part. Only the active labels $E_+=\\{i:w_i>0\\}$ matter. Let $m=|\\phi(E_+)|$. Write $A_{\\phi,w}$ for this squarefree operator. A zero output is allowed; otherwise its coefficient array has rank $d$.\n\n\\begin{theorem}[Weighted set-map classification]\\label{thm:classification}\nThe following are equivalent:\n\\begin{enumerate}\n\\item For every strong $\\T_q$ rank-$d$ input, $A_{\\phi,w}$ is zero or strong over $\\T_q$.\n\\item For every strong $\\T_q$ rank-$d$ input, $A_{\\phi,w}$ is zero or weak over $\\T_q$.\n\\item Either $\\phi|_{E_+}$ is injective, or $m\\le d+1$.\n\\end{enumerate}\nWhen condition 3 fails, failure can be witnessed by a positive uniform input on the original ground set, with every genuine input inequality strict.\n\\end{theorem}\n\\begin{proof}\nCondition 1 implies 2. If the active map is injective, the operation is deletion, positive rescaling, and relabeling; strong representability is preserved, unless every rank-$d$ term is deleted. If $m\\le d+1$, any nonzero rank-$d$ array on these labels is strong: for $m=d$ it has at most one basis, and for $m=d+1$ every strong relation has only its two identical terms. Its support is a matroid, since complements of the possible bases are the nonloops of a rank-one matroid. For $m<d$ the squarefree output is zero. This proves 3 implies 1.\n\nSuppose instead that $\\phi$ has a collision and $m\\ge d+2$. Choose two active labels in one fiber and call them $5,6$. Choose four other distinct fibers and representatives $1,2,3,4$. Choose a further $d-3$ distinct fibers with representatives $C$. The total number of used output labels is $d+2$, so these choices are possible.\n\nApply Theorem~\\ref{thm:explicit} on the entire original ground set with this distinguished configuration. Rescale each active variable by $1/w_i$. Additionally rescale every active variable outside the chosen $d+3$ labels by a parameter $\\lambda>0$. Input positivity and strictness are preserved. In the limit $\\lambda\\to0$, the six output coefficients of the selected weak relation converge to \\eqref{eq:six}; contributions involving any additional element of the same output fibers vanish. Inactive variables contribute nothing. The limiting output relation fails by at least $L/2$, so it still fails for all sufficiently small positive $\\lambda$. This contradicts condition 2 and proves necessity. Every input coefficient remains positive for such $\\lambda$, as asserted.\n\\end{proof}\n\n\\begin{remark}[Why the squarefree qualifier is necessary]\nThe low-output-count exception concerns rank-$d$ arrays on at most $d+1$ labels. The full polynomial may have repeated monomials and additional weak tests. Theorem~\\ref{thm:classification} is not a classification of all full-polynomial linear preservers, nor of arbitrary matrices of linear forms.\n\\end{remark}\n\n\\begin{corollary}[A guaranteed rule for repeated squarefree merging]\nStart with positive weak data on $U_{d,n}$. Perform $k$ binary identifications, taking the squarefree shadow after each, and assume the number of labels remains at least $d$. The resulting array is weak over $\\T_p$ for\n\\[\n p=2^k\\max\\{q,1\\}.\n\\]\nFor $d=2$ the better parameter $\\max\\{q,1\\}$ suffices after any number of such steps.\n\\end{corollary}\n\\begin{proof}\nEach squarefree shadow remains positive on the new uniform support, and its weak tests are among the full output weak tests just proved. Iterate Theorem~\\ref{thm:repair}; in degree two iterate $r\\mapsto\\max\\{r,1\\}$ instead.\n\\end{proof}\nThe exponential bound is a safe composition bound, not an optimal cost for a partition or for multiple merges.", "source_section": "9", "verification_method": "standalone_manuscript", "verification_status": "source_text", "artifact_paths": ["paper/Ilyoxune_Aggregation_Theorem.pdf", "paper/Ilyoxune_Aggregation_Theorem.tex"], "payload_json": "{\"content_format\": \"latex\", \"section_number\": \"9\"}"}
10
+ {"id": "section-10", "record_type": "manuscript_section", "title": "A robust Lorentzian obstruction to strong compatibility", "content": "\\section{A robust Lorentzian obstruction to strong compatibility}\\label{sec:lorentzian}\nThe preceding repair results distinguish weak and strong output. This section gives an independent explicit obstruction: even a Lorentzian cubic can fail strong compatibility in the same parameter that controls its local quadratic geometry.\n\nFor a positive multi-affine cubic on six labels, we use the standard Lorentzian criterion: each quadratic derivative has a Hessian with at most one positive eigenvalue. Uniform support supplies the required M-convexity. This is the usual derivative characterization in \\cite{BH}. The active Hessian omits the zero row and column of the differentiated variable. Signature $(1,4)$ of all six active Hessians implies membership in the relative interior of the Lorentzian $U_{3,6}$ stratum.\n\n\\begin{theorem}[Interior Lorentzian obstruction]\\label{thm:lorentz}\nPartition $[6]$ as $P=\\{1,3\\}$, $Q=\\{2,4,6\\}$, and $\\{s\\}=\\{5\\}$. For $t>1$ and $0<\\delta<1$, put\n\\[\n a=3-\\delta,\\qquad h=4-t^{-2}-\\delta.\n\\]\nAssign the twenty cubic coefficients by the following exhaustive composition classes:\n\\begin{center}\n\\begin{tabular}{lrrrrrr}\n\\toprule\nClass & $PPs$ & $QQQ$ & $PPQ$ & $sQQ$ & $PsQ$ & $PQQ$\\\\\nNumber & 1 & 1 & 3 & 3 & 6 & 6\\\\\nCoefficient & $a$ & $h$ & 1 & 1 & $t$ & $t$\\\\\\bottomrule\n\\end{tabular}\n\\end{center}\nEvery active derivative Hessian has signature $(1,4)$. The strong relation indexed by $I=P$ and $J=Q\\cup\\{s\\}$ has product list $(ah,1,1,1)$. Therefore any parameter $p$ containing all Lorentzian $U_{3,6}$ coefficient arrays among strong $\\T_p$ arrays must satisfy\n\\[\n p\\ge\\log_3 12.\n\\]\n\\end{theorem}\n\\begin{proof}\nAll coefficients are positive. Symmetry reduces the six active Hessians to three types; the quotient matrices below use orthonormal constant vectors on their symmetry blocks.\n\nFor differentiation in $s$, the quotient on $P,Q$ is\n\\[\n \\begin{pmatrix}a&\\sqrt6t\\\\\\sqrt6t&2\\end{pmatrix}.\n\\]\nIts determinant $2a-6t^2$ is negative, giving one positive and one negative eigenvalue. The other three modes have eigenvalues $-a,-1,-1$.\n\nFor differentiation in a member of $P$, the symmetry blocks have sizes $1,1,3$, and the quotient is\n\\[\n \\begin{pmatrix}\n 0&a&\\sqrt3\\\\ a&0&\\sqrt3t\\\\ \\sqrt3&\\sqrt3t&2t\n \\end{pmatrix}.\n\\]\nIts determinant is $2at\\delta>0$ and the sum of its principal two-by-two minors is $-a^2-3-3t^2<0$. Thus its three nonzero eigenvalues have one positive and two negative signs: their product is positive, while the all-positive possibility is excluded by the negative sum of pairwise products. The other two modes are $-t,-t$.\n\nFor differentiation in a member of $Q$, the symmetry blocks have sizes $2,2,1$, and the quotient is\n\\[\n \\begin{pmatrix}\n 1&2t&\\sqrt2t\\\\ 2t&h&\\sqrt2\\\\ \\sqrt2t&\\sqrt2&0\n \\end{pmatrix}.\n\\]\nIts determinant is $2t^2\\delta>0$ and the sum of its principal two-by-two minors is $h-6t^2-2<0$. Again it has one positive and two negative eigenvalues. The other two modes are $-1,-h$. This proves signature $(1,4)$ in every case.\n\nIn the specified strong relation, the $s$ term is $c_{PPs}c_{QQQ}=ah$; each of the other three terms is $c_{PPQ}c_{sQQ}=1$. Strong nullity requires $(ah)^{1/p}\\le3$, so $p\\ge\\log_3(ah)$. Let $\\delta\\downarrow0$ and $t\\to\\infty$ to obtain $p\\ge\\log_3 12$.\n\\end{proof}\n\n\\begin{corollary}[A robust four-level integer certificate]\\label{cor:lorentzbox}\nTake $t=2$, $\\delta=1/4$, and multiply every coefficient by four. The six coefficient classes become\n\\[\n 11,\\quad14,\\quad4,\\quad4,\\quad8,\\quad8.\n\\]\nThis Lorentzian array fails strong $\\T_2$; moreover every coefficient perturbation of absolute size at most $1/1000$ retains both properties within the multi-affine coefficient space.\n\\end{corollary}\n\\begin{proof}\nThe strong product list is $(154,16,16,16)$, and $154>144=(4+4+4)^2$ proves failure. Exact active-Hessian determinants, in variable order, are\n\\[\n 11264,\\ 7168,\\ 11264,\\ 7168,\\ 52096,\\ 7168.\n\\]\nEach maximum absolute row sum is at most $35$. All eigenvalues therefore have absolute value at most $35$, while the determinant product implies every absolute eigenvalue is at least $7168/35^4$. A perturbation of the twenty coefficients by at most $\\eps=1/1000$ changes each active Hessian by operator norm at most $4\\eps$, because each row has only four off-diagonal entries. The exact inequality\n$4/1000<7168/35^4$ prevents any eigenvalue from crossing zero, so the active signatures remain $(1,4)$.\n\nThe worst permitted strong-product comparison still has positive squared excess:\n\\[\n (11-\\eps)(14-\\eps)-9(4+\\eps)^2\n =10-97\\eps-8\\eps^2>0.\n\\]\nAll coefficients remain positive. This gives the stated explicit relative open box.\n\\end{proof}\nThis is a lower-bound theorem for a distinct inclusion problem. It does not establish finiteness or a matching upper bound for the Lorentzian-to-strong parameter.", "source_section": "10", "verification_method": "standalone_manuscript", "verification_status": "source_text", "artifact_paths": ["paper/Ilyoxune_Aggregation_Theorem.pdf", "paper/Ilyoxune_Aggregation_Theorem.tex"], "payload_json": "{\"content_format\": \"latex\", \"section_number\": \"10\"}"}
11
+ {"id": "section-11", "record_type": "manuscript_section", "title": "Concrete certificates and verification", "content": "\\section{Concrete certificates and verification}\\label{sec:verification}\nThe package contains the complete source, an executable verifier, and its JSON results. Universal real-parameter statements are established by the proofs above. Finite computations are certificates for selected constructions and checks against transcription mistakes.\n\n\\subsection{A strict rational witness at $q=1$}\nOn $U_{3,6}$ take root coefficients $a_{135}=199/100$, $a_{126}=a_{346}=1/4$, and all others one; rescale label six by $s=1/4$. Lemma~\\ref{lem:uniform} proves strict strong input. The relevant output coefficients are\n\\[\n g_{12}=g_{34}=17/16,\\quad\n g_{14}=g_{23}=g_{24}=5/4,\\quad\n g_{13}=56/25.\n\\]\nThe failed triangle defect, with no numerical roots required, is\n\\[\n \\frac{56}{25}\\frac54-\\left(\\frac54\\right)^2-\n \\left(\\frac{17}{16}\\right)^2=\\frac{139}{1280}>0.\n\\]\nThis example is convenient for an independent manual check of strictness, merging, and exact rational failure.\n\n\\subsection{A strict integer witness at $q=2$}\nAn explicit integer construction is\n\\[\n c_{135}=4095,\\quad c_{126}=c_{346}=8,\\quad\n c_S=64\\ (6\\in S\\text{ otherwise}),\\quad\n c_S=1024\\ (6\\notin S\\text{ otherwise}).\n\\]\nIts six merged coefficients are $g_{12}=g_{34}=1032$, $g_{14}=g_{23}=g_{24}=1088$, and $g_{13}=4159$. Failure is certified by\n\\[\n 4159\\cdot1088-(1088+1032)^2=30592>0.\n\\]\nThe input is obtained from Lemma~\\ref{lem:uniform} with root peak $\\sqrt{4-1/1024}$ and low $1/\\sqrt8$, followed by label-six scaling $1/16$ and global scaling $1024$. Exact enumeration provides a second certificate.\n\n\\subsection{What is checked}\nThe delivered scripts separate exact certificates from numerical checks.\n\\begin{itemize}\n\\item The core verifier enumerates all strong relations in eight rank/size configurations, certifies the rational and integer examples, and checks all six normalized output cases as formal polynomial identities. It also certifies 567 boundary-case relations exactly and checks 264 numerical instances containing 11,424 normalized weak relations.\n\\item The congruence verifier checks three formal tensor identities, 16 exact rational matrix instances, 128 numerical substitution/composition instances at 90 digits, the minimal corank example, the two-peak classification in selected ranks and sizes, and the integer levels $6,14,23$.\n\\item The strong-parameter verifier certifies every one of 225 indexed input relations for each of two integer witnesses, then checks the output twentieth-power comparisons entirely with integers and fractions.\n\\item The Lorentzian verifier proves twelve derivative quotient identities symbolically and checks 372 principal minors exactly at the rational examples. It also verifies the strict integer example and every stated robustness constant.\n\\end{itemize}\nAll delivered checks pass. Universal real-parameter statements are established by the written proofs, rather than inferred from samples. The scripts are reproducibility certificates, not a proof-assistant formalization.", "source_section": "11", "verification_method": "standalone_manuscript", "verification_status": "source_text", "artifact_paths": ["paper/Ilyoxune_Aggregation_Theorem.pdf", "paper/Ilyoxune_Aggregation_Theorem.tex"], "payload_json": "{\"content_format\": \"latex\", \"section_number\": \"11\"}"}
12
+ {"id": "section-12", "record_type": "manuscript_section", "title": "Interpretation, limits, and the next mathematical target", "content": "\\section{Interpretation, limits, and the next mathematical target}\nThe decisive structural distinction is between homogeneous rescaling and additive aggregation. Rescaling acts by one common multiplier in each relation. Aggregation sums coefficients, and a common-index slice becomes a pointwise aggregate of two metrics after a shared normalization. Passing to a square root restores the Ptolemy property of that aggregate; this is the source of the factor two in the repair.\n\nThe exact quadratic theorem allows arbitrary nonnegative linear maps and their compositions. The cardinality table identifies the complete same-parameter boundary for a full merge. Above that boundary, explicit uniform obstructions and a constructive weak repair coexist. The two-peak family resolves weak versus strong feasibility in a complete parameter region, while the Lorentzian family exhibits a different robust obstruction to strong compatibility.\n\nFor automated algebra, the practical consequence is a mathematically sound composition rule: a system may lower its claimed output certificate from strong $\\T_q$ to weak $\\T_{2\\max(q,1)}$ after one merge of matroid data. If the same parameter or a strong output is required, new relation checks remain necessary. The paper supplies constant-size regression witnesses and a general operator rejection criterion. It does not measure an AI speedup or establish recursive self-improvement of a model.\n\nA focused remaining question is the exact function $P^{\\rm str}_{d,n}(q)$ for $q>1$. The current bounds differ by at most two and have the exact asymptotic factor. A second open extension is a comparable repair theorem when the initial input already has repeated coordinates, rather than being multi-affine. A third is higher-degree classification for arbitrary nonnegative linear substitutions rather than weighted set maps; the quadratic case is completely covered by Theorem~\\ref{thm:congruence}. None is silently included in the proved assertions.\n\n\\appendix", "source_section": "12", "verification_method": "standalone_manuscript", "verification_status": "source_text", "artifact_paths": ["paper/Ilyoxune_Aggregation_Theorem.pdf", "paper/Ilyoxune_Aggregation_Theorem.tex"], "payload_json": "{\"content_format\": \"latex\", \"section_number\": \"12\"}"}
13
+ {"id": "section-a", "record_type": "manuscript_section", "title": "Why merging matroid support is M-convex", "content": "\\section{Why merging matroid support is M-convex}\n\\begin{lemma}\\label{lem:support}\nThe image of a matroid's basis indicators under one coordinate merge is M-convex.\n\\end{lemma}\n\\begin{proof}\nLet $\\phi:E\\to H$ merge $u,v$ into $z$, leaving other labels distinct. Define the integral monotone submodular function $f(A)=r_M(\\phi^{-1}(A))$. Its basis images are exactly the integer vectors\n\\[\n x\\ge0,\\qquad x(H)=d,\\qquad x(A)\\le f(A)\\quad(A\\subseteq H).\n\\]\nNecessity is immediate. Conversely, every unmerged $x_i\\le1$. Let $S$ be the unmerged labels with $x_i=1$, and put $k=x_z\\le2$. The inequalities imply that $S$ is independent and $r_M(S\\cup\\{u,v\\})\\ge d$. Thus $S$ extends to a basis using exactly $k=d-|S|$ elements of $\\{u,v\\}$, as required.\n\nWe prove symmetric exchange for these vectors. Let $x_i>y_i$. Tight sets for $x$ are closed under unions and intersections by submodularity, as are tight sets for $y$. Let $U$ be the union of all $x$-tight sets avoiding $i$ (including the empty set), and $V$ the intersection of all $y$-tight sets containing $i$ (including $H$). Then $i\\in V\\setminus U$. An exchange $x-e_i+e_j$ preserves all upper inequalities if $j\\notin U$; $y+e_i-e_j$ does so if $j\\in V$. Integer slack makes every inequality that is not tight harmless.\n\nTightness and submodularity give\n\\[\n x(V\\setminus U)\\le f(U\\cup V)-f(U)\n \\le f(V)-f(U\\cap V)\\le y(V\\setminus U).\n\\]\nSince $x_i>y_i$, some $j\\in V\\setminus U$ has $x_j<y_j$. Both exchanged vectors are therefore feasible nonnegative integer vectors. This is M-convexity.\n\\end{proof}", "source_section": "A", "verification_method": "standalone_manuscript", "verification_status": "source_text", "artifact_paths": ["paper/Ilyoxune_Aggregation_Theorem.pdf", "paper/Ilyoxune_Aggregation_Theorem.tex"], "payload_json": "{\"content_format\": \"latex\", \"section_number\": \"A\"}"}
data/theorems.jsonl ADDED
@@ -0,0 +1,25 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"id": "theorem-01-thm-main", "record_type": "theorem", "title": "Principal results", "content": "\\begin{theorem}[Principal results]\\label{thm:main}\nLet $q>0$.\n\\begin{enumerate}\n\\item For any symmetric nonnegative matrix $C$ satisfying all $\\T_q$ quartet inequalities and every nonnegative rectangular matrix $A$, the matrix $A^{\\mathsf T}CA$ satisfies all quartet inequalities at the exact optimal universal parameter $\\max\\{q,1\\}$. This controls the full normalized quadratic polynomial under arbitrary nonnegative linear substitutions.\n\\item One merge of rank-$d$ matroid data on $n$ labels preserves strong representability at the same parameter precisely in the regimes in Table~\\ref{tab:threshold}. The same classification holds for universal strong-to-weak preservation.\n\\item Every weak $\\T_q$ matroid input, including zero coefficients, has full normalized weak $\\T_{2\\max(q,1)}$ output after one merge. The output parameter is exactly optimal for $0<q\\le1$, and its asymptotic factor two is optimal.\n\\item For every $d\\ge3,n\\ge d+3$, explicit strict strong uniform inputs fail after one merge at parameter $q$. At $q=2$ a quantitative certificate forces the universal strong-input repair parameter above $20/9$.\n\\item Weighted set identifications followed by squarefree projection preserve all strong rank-$d$ arrays, $d\\ge3$, exactly when their active map is injective or has at most $d+1$ output labels.\n\\item There is a family in the relative interior of the Lorentzian $U_{3,6}$ stratum forcing every universal Lorentzian-to-strong parameter to be at least $\\log_3 12$.\n\\end{enumerate}\n\\end{theorem}", "source_section": "1", "verification_method": "written_mathematical_proof", "verification_status": "proved_in_manuscript_sections", "artifact_paths": ["paper/Ilyoxune_Aggregation_Theorem.pdf", "paper/Ilyoxune_Aggregation_Theorem.tex"], "payload_json": "{\"content_format\": \"latex\", \"latex_label\": \"thm:main\", \"proof_included_in_record\": false}"}
2
+ {"id": "lemma-02-lem-basic", "record_type": "lemma", "title": "Monotonicity and rescaling", "content": "\\begin{lemma}[Monotonicity and rescaling]\\label{lem:basic}\nIf a list is null at $q$, it is null at every $p\\ge q$. Multiplying coefficients by $\\lambda\\prod_{i\\in S}d_i$, for $\\lambda,d_i>0$, preserves all strong and weak relations and their strictness.\n\\end{lemma}\n\n\\begin{proof}\nFor monotonicity put $\\theta=q/p\\le1$ and use $(\\sum a_i)^\\theta\\le\\sum a_i^\\theta$ for nonnegative $a_i$. In a fixed strong relation, every product acquires the common factor $\\lambda^2\\prod_{i\\in I}d_i\\prod_{j\\in J}d_j$. The same argument covers weak relations.\n\\end{proof}", "source_section": "2", "verification_method": "written_mathematical_proof", "verification_status": "proof_included", "artifact_paths": ["paper/Ilyoxune_Aggregation_Theorem.pdf", "paper/Ilyoxune_Aggregation_Theorem.tex"], "payload_json": "{\"content_format\": \"latex\", \"latex_label\": \"lem:basic\", \"proof_included_in_record\": true}"}
3
+ {"id": "lemma-03-lem-minkowski", "record_type": "lemma", "title": "Aggregation of triangles", "content": "\\begin{lemma}[Aggregation of triangles]\\label{lem:minkowski}\nLet $X_h,Y_h,Z_h\\ge0$ satisfy all triangle inequalities for each $h$. Put $r=\\max\\{q,1\\}$. Then\n\\[\n (\\sum_hX_h^q)^{1/r},\\quad\n (\\sum_hY_h^q)^{1/r},\\quad\n (\\sum_hZ_h^q)^{1/r}\n\\]\nsatisfy all triangle inequalities.\n\\end{lemma}\n\n\\begin{proof}\nFor $q\\ge1$ this is Minkowski's inequality, applied to $X_h\\le Y_h+Z_h$, and its cyclic analogues. For $q\\le1$, $X_h^q\\le(Y_h+Z_h)^q\\le Y_h^q+Z_h^q$; sum over $h$.\n\\end{proof}", "source_section": "2", "verification_method": "written_mathematical_proof", "verification_status": "proof_included", "artifact_paths": ["paper/Ilyoxune_Aggregation_Theorem.pdf", "paper/Ilyoxune_Aggregation_Theorem.tex"], "payload_json": "{\"content_format\": \"latex\", \"latex_label\": \"lem:minkowski\", \"proof_included_in_record\": true}"}
4
+ {"id": "lemma-04-lem-snowflake", "record_type": "lemma", "title": "Square roots of metrics are Ptolemaic", "content": "\\begin{lemma}[Square roots of metrics are Ptolemaic]\\label{lem:snowflake}\nFor every metric $D$ and four points,\n\\begin{equation}\\label{eq:snowflake}\n \\sqrt{D_{13}D_{24}}\\le\n \\sqrt{D_{12}D_{34}}+\\sqrt{D_{14}D_{23}}.\n\\end{equation}\n\\end{lemma}\n\n\\begin{proof}\nWrite the successive side distances as $a,b,c,e$, and the diagonals as $f,g$. The triangle inequalities give\n$f\\le\\min(a+b,c+e)$ and $g\\le\\min(a+e,b+c)$.\nBy relabeling around the four-cycle the two chosen minima can be written $a+b$ and $a+e$: the two chosen paths always share a side. Their being minima implies $c\\ge a+|b-e|$. It suffices to prove\n\\[\n (a+b)(a+e)\\le(\\sqrt{ac}+\\sqrt{be})^2.\n\\]\nAfter cancellation this is $a(a+b+e-c)\\le2\\sqrt{abce}$. The left side decreases with $c$ and the right side increases. At $c=a+|b-e|$, assume $b\\ge e$ by reversing the cycle; the required inequality reduces to $ae\\le b(a+b-e)$, which follows from\n$b(a+b-e)-ae=(b-e)(a+b)\\ge0$. Zero distances follow by continuity. This proves \\eqref{eq:snowflake}.\n\\end{proof}", "source_section": "2", "verification_method": "written_mathematical_proof", "verification_status": "proof_included", "artifact_paths": ["paper/Ilyoxune_Aggregation_Theorem.pdf", "paper/Ilyoxune_Aggregation_Theorem.tex"], "payload_json": "{\"content_format\": \"latex\", \"latex_label\": \"lem:snowflake\", \"proof_included_in_record\": true}"}
5
+ {"id": "theorem-05-thm-congruence", "record_type": "theorem", "title": "Sharp nonnegative congruence", "content": "\\begin{theorem}[Sharp nonnegative congruence]\\label{thm:congruence}\nIf $C\\in\\mathcal C_q$ and $A$ is any entrywise nonnegative rectangular matrix, then\n\\[\n A^{\\mathsf T}CA\\in\\mathcal C_{\\max(q,1)}.\n\\]\nThe parameter $\\max\\{q,1\\}$ is optimal as a universal rule. In particular $\\mathcal C_q$ is preserved by every nonnegative congruence when $q\\ge1$.\n\\end{theorem}\n\n\\begin{proof}\nSet $B=A^{\\mathsf T}CA$, and fix any output quartet $a,b,c,d$, possibly repeated. For each ordered source quadruple define\n\\[\n W_{ijkl}=A_{ia}A_{jb}A_{kc}A_{\\ell d}.\n\\]\nThe three output products expand with exactly the same nonnegative weights:\n\\begin{align*}\n B_{ab}B_{cd}&=\\sum_{i,j,k,\\ell}W_{ijkl}C_{ij}C_{k\\ell},\\\\\n B_{ac}B_{bd}&=\\sum_{i,j,k,\\ell}W_{ijkl}C_{ik}C_{j\\ell},\\\\\n B_{ad}B_{bc}&=\\sum_{i,j,k,\\ell}W_{ijkl}C_{i\\ell}C_{jk}.\n\\end{align*}\nFor each source quadruple, take the $q$th roots of its three weighted terms. They satisfy all triangle inequalities. Lemma~\\ref{lem:minkowski} proves the three required inequalities at $r=\\max\\{q,1\\}$ after summing. No positivity, invertibility, or restriction on the shape of $A$ is used.\n\nTo force $p\\ge q$, take the absolute-minor $q$-power array of the real columns $(1,0),(0,1),(1,1),(1,-1)$ and choose the identity substitution. Its extremal quartet products are $(2^q,1,1)$.\nTo force $p\\ge1$, start with $F=x_1x_2$, valid at every $q$, and substitute\n\\[\n x_1=y_1+z,\\qquad x_2=y_2+z.\n\\]\nThe output is $y_1y_2+y_1z+y_2z+z^2$. Its normalized repeated-$z$ quartet products are exactly $(2,1,1)$, requiring $2^{1/p}\\le2$, or $p\\ge1$. Both lower bounds are attained by the stated upper bound.\n\\end{proof}", "source_section": "3", "verification_method": "written_mathematical_proof", "verification_status": "proof_included", "artifact_paths": ["paper/Ilyoxune_Aggregation_Theorem.pdf", "paper/Ilyoxune_Aggregation_Theorem.tex"], "payload_json": "{\"content_format\": \"latex\", \"latex_label\": \"thm:congruence\", \"proof_included_in_record\": true}"}
6
+ {"id": "lemma-06-lem-quadraticsupport", "record_type": "lemma", "title": "Support is automatic", "content": "\\begin{lemma}[Support is automatic]\\label{lem:quadraticsupport}\nThe nonzero quadratic coefficient support of a matrix in $\\mathcal C_q$ is M-convex.\n\\end{lemma}\n\n\\begin{proof}\nDelete zero rows. On the remaining indices define $i\\sim j$ when $i=j$ or $C_{ij}=0$. Suppose $C_{ij}=C_{jk}=0$ but $C_{ik}>0$. Since row $j$ is nonzero, choose $\\ell$ with $C_{j\\ell}>0$. The quartet $(i,k,j,\\ell)$ has exactly one positive product, a contradiction. Thus $\\sim$ is an equivalence relation. If $i\\ne j$ are in one class, choose a positive neighbor of $j$; the quartet $(i,i,j,\\ell)$ forces $C_{ii}=0$. Positive diagonal entries therefore occur only in singleton classes.\n\nThe support consists exactly of degree-two integer vectors having total at most one on every zero-diagonal class, and total at most two on each positive-diagonal singleton. These partition-capacity bases satisfy symmetric exchange. Indeed, for $x_i>y_i$, first choose a deficit coordinate in the same block if one exists. Otherwise $x$ has greater total than $y$ on that block, so choose a deficit coordinate in a block where $x$ has smaller total. The exchange respects both capacities for both vectors. Hence the support is M-convex.\n\\end{proof}", "source_section": "3", "verification_method": "written_mathematical_proof", "verification_status": "proof_included", "artifact_paths": ["paper/Ilyoxune_Aggregation_Theorem.pdf", "paper/Ilyoxune_Aggregation_Theorem.tex"], "payload_json": "{\"content_format\": \"latex\", \"latex_label\": \"lem:quadraticsupport\", \"proof_included_in_record\": true}"}
7
+ {"id": "corollary-07-exact-quadratic-substitution-calculus", "record_type": "corollary", "title": "Exact quadratic substitution calculus", "content": "\\begin{corollary}[Exact quadratic substitution calculus]\nWrite $F(x)=\\tfrac12x^{\\mathsf T}Cx$ with normalized quadratic coefficient matrix $C\\in\\mathcal C_q$. For every nonnegative linear substitution $x=Ay$, the full normalized output matrix is exactly $A^{\\mathsf T}CA$ and if nonzero, its weak and strong parameter is universally repaired by $p=\\max\\{q,1\\}$. No parameter loss accumulates under any finite composition once the parameter is at least one.\n\\end{corollary}\n\n\\begin{proof}\nThe matrix identity follows by substitution. The support lemma supplies M-convexity. All rank-two strong relations have at most three terms, so the quartet conditions give both weak and strong representations. A composition of nonnegative matrices is nonnegative; apply Theorem~\\ref{thm:congruence} to their product.\n\\end{proof}", "source_section": "3", "verification_method": "written_mathematical_proof", "verification_status": "proof_included", "artifact_paths": ["paper/Ilyoxune_Aggregation_Theorem.pdf", "paper/Ilyoxune_Aggregation_Theorem.tex"], "payload_json": "{\"content_format\": \"latex\", \"latex_label\": \"\", \"proof_included_in_record\": true}"}
8
+ {"id": "lemma-08-lem-uniform", "record_type": "lemma", "title": "Three exceptional bases", "content": "\\begin{lemma}[Three exceptional bases]\\label{lem:uniform}\nFor $1<b<2$ and $0<e\\le1/2$, set $a_U=b$, $a_{Z_1}=a_{Z_2}=e$, and $a_S=1$ on every other $d$-subset of an $n$-element ground set, $n\\ge d+3$. For every $q>0$, the coefficient vector $c_S=a_S^q$ is strict strong over $\\T_q$ on $U_{d,n}$. Every genuine root-polygon slack is at least\n\\begin{equation}\\label{eq:slack}\n \\delta=\\min\\{e^2,2-b\\}.\n\\end{equation}\n\\end{lemma}\n\n\\begin{proof}\nAfter taking roots, each strong term is a product of two $a$'s. In a relation with $m\\ge3$ terms, each exceptional coefficient can occur at most once among the factors: two different first factors are distinct, two different second factors are distinct, and a first factor cannot equal a second factor when $m\\ge3$. Every term is at most two; a term containing a low coefficient is at most $be<1$.\n\nFor $m=3$, factors from different terms intersect in $d-1$ elements. The three exceptional bases intersect pairwise in $d-2$ elements. Consequently any two exceptional factors occurring in the relation must occur in the same term. At least two terms therefore equal one.\n\nFor $m=4$, if fewer than three terms are exceptional there are again two unit terms. If all three exceptional bases occur, the union $I\\cup J$ has size $d+3$ and equals $C\\cup[6]$. Complement each basis inside this union. The dual relation has rank three and its exceptional triples are a peak at $246$ and lows at $125,345$. The permutation $(12)(34)(56)$ takes them to the original peak at $135$ and lows at $126,346$. The rank-three four-term products have at least two unit terms. To see this directly, complementary products at the sparse limit have possible lists\n\\[\n (1,1,1,1),\\ (0,1,1,1),\\ (0,0,1,1),\\ (1,1,1,2),\\ (0,1,1,2).\n\\]\nThe two zeros can occur together only when $I=12$ or $34$, in which case the peak product cannot occur. Restoring positive lows and replacing the peak root by $b$ retains the two unit terms.\n\nFor $m\\ge5$, there are at most three exceptional terms, so at least $m-3\\ge2$ unit terms remain. Thus in every case a peak term has slack at least $2-b$. Without a peak term the largest term is at most one, two unit terms remain, and at least one additional term is positive and at least $e^2$. This proves \\eqref{eq:slack} and strictness.\n\\end{proof}", "source_section": "4", "verification_method": "written_mathematical_proof", "verification_status": "proof_included", "artifact_paths": ["paper/Ilyoxune_Aggregation_Theorem.pdf", "paper/Ilyoxune_Aggregation_Theorem.tex"], "payload_json": "{\"content_format\": \"latex\", \"latex_label\": \"lem:uniform\", \"proof_included_in_record\": true}"}
9
+ {"id": "theorem-09-thm-explicit", "record_type": "theorem", "title": "Explicit all-rank obstruction", "content": "\\begin{theorem}[Explicit all-rank obstruction]\\label{thm:explicit}\nFor every $q>0$, define\n\\begin{equation}\\label{eq:parameters}\n T=2^q,\\quad D=2^{-q}-4^{-q},\\quad\n s=\\frac{qD}{64(q+2)},\\quad\n L=\\frac{3D^2}{128(q+2)},\\quad b=2-\\frac{L}{16}.\n\\end{equation}\nUse the root array of Lemma~\\ref{lem:uniform} with $e=1/4$ and put\n\\begin{equation}\\label{eq:explicit}\n c_S=a_S^q s^{\\one_{6\\in S}}.\n\\end{equation}\nFor every $d\\ge3,n\\ge d+3$, this is positive strict strong on $U_{d,n}$, whereas merging $5,6$ fails a squarefree weak relation by at least $L/2$ in root lengths.\n\\end{theorem}\n\n\\begin{proof}\nSince $0<D\\le1/4$, we have $L<1$, $1<b<2$, and $0<s\\le\\min(1,q)$. Lemmas~\\ref{lem:uniform} and \\ref{lem:basic} prove the input assertion.\n\nIn the output use common index $C\\cup\\{z\\}$ and quartet $1,2,3,4$. Omitting the common $C$ from notation, the six coefficients are\n\\begin{equation}\\label{eq:six}\n g_{12}=g_{34}=1+4^{-q}s,\\qquad\n g_{14}=g_{23}=g_{24}=1+s,\\qquad g_{13}=b^q+s.\n\\end{equation}\nTemporarily replace $b$ by $2$ and define\n\\[\n H(t)=((T+t)(1+t))^{1/q}-(1+t)^{2/q}-(1+4^{-q}t)^{2/q}.\n\\]\nThen $H(0)=0$ and $H'(0)=2D/q$. Put $r_0=1/q$. For $0\\le t\\le\\min(1,q)$, the three positive expressions defining $H$ are at most $2\\exp(2),\\exp(2),\\exp(2)$. Logarithmic differentiation bounds the absolute second derivative of each expression by its value times $4r_0^2+2r_0$. Since $\\exp(2)<8$,\n\\[\n |H''(t)|<64r_0(2r_0+1)=M.\n\\]\nThe selected $s$ satisfies $Ms=r_0D$. Taylor's integral remainder gives\n\\[\n H(s)\\ge2r_0Ds-\\tfrac12Ms^2\n \\ge\\tfrac32r_0Ds=L.\n\\]\nOnly the first expression changes when the peak root decreases from $2$ to $b$. For $1\\le v\\le2$ its derivative is\n\\[\n \\frac{\\partial}{\\partial v}((v^q+s)(1+s))^{1/q}\n =(1+s)^{r_0}(1+s/v^q)^{r_0-1}<8.\n\\]\nIndeed, if $r_0-1<0$ its second factor is at most one; otherwise the product is at most $\\exp(2r_0s)\\le\\exp(2)<8$. Decreasing $v$ by $L/16$ loses at most $L/2$. Thus\n\\[\n (g_{13}g_{24})^{1/q}-(g_{14}g_{23})^{1/q}-(g_{12}g_{34})^{1/q}\\ge L/2>0.\n\\]\nExtra labels never enter these six coefficients, proving the claim in every $d,n$.\n\\end{proof}", "source_section": "4", "verification_method": "written_mathematical_proof", "verification_status": "proof_included", "artifact_paths": ["paper/Ilyoxune_Aggregation_Theorem.pdf", "paper/Ilyoxune_Aggregation_Theorem.tex"], "payload_json": "{\"content_format\": \"latex\", \"latex_label\": \"thm:explicit\", \"proof_included_in_record\": true}"}
10
+ {"id": "corollary-10-cor-robust", "record_type": "corollary", "title": "Explicit robustness independent of dimension", "content": "\\begin{corollary}[Explicit robustness independent of dimension]\\label{cor:robust}\nLet $c$ be \\eqref{eq:explicit} and put $\\eta=L/2048$. Every positive array $c'$ satisfying\n\\[\n 1-\\eta\\le(c'_S/c_S)^{1/q}\\le1+\\eta\\qquad\\text{for every }S\n\\]\nis still strict strong and still fails after the same merge. Its normalized input root-polygon slack is at least $3L/256$, and its output defect is at least $3L/32$.\n\\end{corollary}\n\n\\begin{proof}\nFactor the common diagonal-rescaling multiplier from each input relation. Each root product changes by a factor in $[(1-\\eta)^2,(1+\\eta)^2]$. Its original maximum is at most two and its slack at least $L/16$. For any chosen side the new slack is at least\n\\[\n (1-\\eta)^2\\frac L{16}-8\\eta\n \\ge\\frac L{64}-\\frac L{256}=\\frac{3L}{256}.\n\\]\nThe same relative bounds hold on the roots of merged coefficient products: sums preserve the coefficient bounds before taking roots. The sum of the two smaller original output root lengths is less than $16$, using the bounds in the preceding proof. The new defect is therefore at least\n\\[\n (1-\\eta)^2\\frac L2-4\\eta\\cdot16\n \\ge\\frac L8-\\frac L{32}=\\frac{3L}{32}.\n\\]\n\\end{proof}", "source_section": "4", "verification_method": "written_mathematical_proof", "verification_status": "proof_included", "artifact_paths": ["paper/Ilyoxune_Aggregation_Theorem.pdf", "paper/Ilyoxune_Aggregation_Theorem.tex"], "payload_json": "{\"content_format\": \"latex\", \"latex_label\": \"cor:robust\", \"proof_included_in_record\": true}"}
11
+ {"id": "theorem-11-thm-repair", "record_type": "theorem", "title": "One-merge repair", "content": "\\begin{theorem}[One-merge repair]\\label{thm:repair}\nLet $c_S>0$ for every $d$-subset of $[n]$, and suppose $c$ is weak over $\\T_q$. For $r=\\max\\{q,1\\}$, the normalized coefficient function \\eqref{eq:rho} of $D_{u,v}F$ is weak over $\\T_{2r}$. If $d=2$, $\\T_r$ suffices. Every larger output parameter also suffices.\n\\end{theorem}\n\n\\begin{proof}\nThe support is the capped simplex described in Section~\\ref{sec:definitions}, so it remains to verify all tests \\eqref{eq:generalweak}. Let $s_0=\\alpha_z$ and let $k_0$ be the multiplicity of $z$ in the quartet. Capacities give $s_0+k_0\\le2$, so there are exactly six cases.\n\n\\textbf{Case $(s_0,k_0)=(0,0)$.} The products are an original input relation. They remain null at $2r\\ge q$ by monotonicity.\n\n\\textbf{Case $(0,1)$.} Write $K$ for the squarefree common index. Each output product is a sum of two products, one using $u$, one using $v$. For example,\n\\[\n c_{Kij}(c_{Kku}+c_{Kkv})=\\sum_{h\\in\\{u,v\\}}c_{Kij}c_{Kkh}.\n\\]\nFor each fixed $h$, the three $q$th roots form a triangle by the input relation on $i,j,k,h$. Lemma~\\ref{lem:minkowski} gives the output triangle at $p=r$, and hence at $2r$.\n\n\\textbf{Case $(0,2)$.} Let $K$ be the common index and $i,j$ the two nonmerged quartet labels. Put $a=c_{Kij}$, $w=c_{Kuv}$, and $b_i=c_{Kiu}+c_{Kiv}$, with $b_j$ defined similarly. The input quartet $i,j,u,v$ gives\n\\[\n (aw)^{1/q}\\le(c_{Kiu}c_{Kjv})^{1/q}+(c_{Kiv}c_{Kju})^{1/q}.\n\\]\nUsing $(x+y)^q\\le2^{r-1}(x^q+y^q)$ for $q\\ge1$, and subadditivity for $q\\le1$, yields\n\\[\n 2aw\\le2^r(c_{Kiu}c_{Kjv}+c_{Kiv}c_{Kju})\\le2^r b_i b_j.\n\\]\nThe output products are $2aw,b_i b_j,b_i b_j$. Thus they form a null list already at $p=r$.\n\n\\textbf{Case $(1,0)$: the only parameter-doubling case.}\nWrite $\\alpha=C+z$, $|C|=d-3$, and let the quartet be $i,j,k,\\ell$. Define\n\\[\n a^h_{ij}=c_{Cijh}^{1/q}\\quad(h=u,v),\\qquad w_i=c_{Ciuv}^{1/q}>0.\n\\]\nApply the input weak relation with common index $C\\cup\\{u\\}$ and quartet $i,j,k,v$. One of its triangle inequalities is\n\\[\n a^u_{ij}w_k\\le a^u_{ik}w_j+a^u_{jk}w_i.\n\\]\nThe analogous statement holds for $v$. Hence\n\\[\n d^h_{ij}=\\frac{a^h_{ij}}{w_iw_j},\\qquad d^h_{ii}=0,\n\\]\nare metrics on the unmerged labels under consideration. Positivity is essential here and is part of the theorem's hypotheses. For $q\\ge1$ their pointwise $\\ell_q$ aggregate\n\\[\n D_{ij}=((d^u_{ij})^q+(d^v_{ij})^q)^{1/q}\n\\]\nis a metric by Minkowski. For $q\\le1$, each $(d^h)^q$ is a metric, so $D_{ij}=(d^u_{ij})^q+(d^v_{ij})^q$ is a metric. In both cases the merged coefficients satisfy\n\\begin{equation}\\label{eq:metricfactor}\n b_{ij}:=c_{Ciju}+c_{Cijv}=(w_iw_j)^qD_{ij}^{r}.\n\\end{equation}\nAt parameter $p=2r$, the root product associated with a pairing of the quartet equals\n\\[\n (b_{ij}b_{k\\ell})^{1/(2r)}\n =(w_iw_jw_kw_\\ell)^{q/(2r)}\\sqrt{D_{ij}D_{k\\ell}}.\n\\]\nThe common positive multiplier cancels. Lemma~\\ref{lem:snowflake} supplies precisely all three required inequalities.\n\n\\textbf{Case $(1,1)$.} Write $\\alpha=C+z$ and use quartet $i,j,k,z$. Set $W_i=c_{Ciuv}$ and $B_{jk}=c_{Cjku}+c_{Cjkv}$. The three output products are\n\\[\n 2W_iB_{jk},\\quad2W_jB_{ik},\\quad2W_kB_{ij}.\n\\]\nFor each fixed $h=u,v$, the products $W_i c_{Cjkh}$, $W_j c_{Cikh}$, $W_k c_{Cijh}$ form an input weak relation with common index $C\\cup\\{h\\}$. Aggregate the two lists using Lemma~\\ref{lem:minkowski} and discard the shared factor two. This proves nullity at $p=r$.\n\n\\textbf{Case $(2,0)$.} The common index has the form $C+2z$. Each coefficient is twice an original coefficient with $u,v$ present. All three products are multiplied by four, and the input weak relation with common index $C\\cup\\{u,v\\}$ applies.\n\nThese six cases exhaust all admissible tests. In degree two, $|\\alpha|=0$, so only the first three cases occur and $p=r$ suffices throughout. Monotonicity proves the assertion for larger $p$.\n\\end{proof}", "source_section": "5", "verification_method": "written_mathematical_proof", "verification_status": "proof_included", "artifact_paths": ["paper/Ilyoxune_Aggregation_Theorem.pdf", "paper/Ilyoxune_Aggregation_Theorem.tex"], "payload_json": "{\"content_format\": \"latex\", \"latex_label\": \"thm:repair\", \"proof_included_in_record\": true}"}
12
+ {"id": "theorem-12-thm-zeros", "record_type": "theorem", "title": "Arbitrary matroid support", "content": "\\begin{theorem}[Arbitrary matroid support]\\label{thm:zeros}\nThe normalized output of one merge of any weak $\\T_q$ matroid coefficient array is weak over $\\T_{2\\max\\{q,1\\}}$. In degree two, $\\T_{\\max\\{q,1\\}}$ suffices, with strong output as well.\n\\end{theorem}\n\n\\begin{proof}\nThe output support is M-convex by Lemma~\\ref{lem:support} below. The five elementary cases in the preceding proof allow zeros unchanged. Consider its critical case $(1,0)$ and fix its quartet. Retain the notation $a^h_{ij}=c_{Cijh}^{1/q}$ and $w_i=c_{Ciuv}^{1/q}$.\n\nIf every $w_i>0$, the preceding metric argument applies directly with pseudometrics: its divisions are valid, and vanishing distances do not affect Minkowski or Lemma~\\ref{lem:snowflake}. Hence assume that at least one weight vanishes.\n\nSuppose some $w_i$ is positive. The weighted triangle inequalities imply that an edge between two zero-weight vertices has $a^h_{ij}=0$: use a positive third vertex. They also imply that for a zero-weight vertex $i$ and positive-weight vertices $j,k$, $a^h_{ij}/w_j=a^h_{ik}/w_k=:A_{i,h}$, by applying both directions of the same inequality. If at least two weights vanish, the output products are either all zero or $(0,B,B)$, so are null at every parameter.\n\nIf exactly one weight, say $w_i$, vanishes, put $M=A_{i,u}^q+A_{i,v}^q$. For positive vertices $j$ we have $b_{ij}=w_j^q M$. On the other three vertices the aggregate metric $D$ in \\eqref{eq:metricfactor} is defined as before, allowing zero distances. If $M=0$ all output products vanish. Otherwise each output root product at parameter $2r$ is a common positive multiple of one of $\\sqrt{D_{jk}},\\sqrt{D_{j\\ell}},\\sqrt{D_{k\\ell}}$. These form a triangle, since square roots of metric distances satisfy the triangle inequality. Lemma~\\ref{lem:snowflake} also holds for pseudometrics by the same proof.\n\nIt remains to treat $w_i=0$ at all four vertices. The input weak relation with common index $C\\cup\\{i\\}$ and quartet $j,k,u,v$ now has two nonzero candidate products, and forces\n\\[\n c_{Ciju}c_{Cikv}=c_{Cijv}c_{Ciku}.\n\\]\nThus nonzero coefficient two-vectors on adjacent edges are proportional. Suppose two disjoint nonzero edge vectors $v_{ij}=(c_{Ciju},c_{Cijv})$ and $v_{k\\ell}$ were not proportional. Every bridge vector $v_{ik},v_{i\\ell},v_{jk},v_{j\\ell}$ is adjacent to both and must be zero. Choose nonzero coordinates $h,g\\in\\{u,v\\}$ to obtain actual matroid bases $C\\cup\\{i,j,h\\}$ and $C\\cup\\{k,\\ell,g\\}$. Basis exchange removing $i$ from the first basis is impossible: exchanging with $k$ or $\\ell$ gives a zero bridge coefficient, and exchanging with $g\\ne h$ gives $c_{Cjuv}=0$. This contradicts the support matroid's basis-exchange axiom.\n\nAll six nonnegative edge vectors therefore lie on one ray (zero vectors allowed). If every vector is zero there is nothing to prove. Otherwise choose a component $h$ positive on that ray. There is a single scalar $\\lambda>0$ such that $b_{ij}=\\lambda c_{Cijh}$ for every edge. The output relation is a rescaled input weak relation, valid already at $q$ and hence at $2r$. This completes all zero cases without a density assumption. Degree two never uses this critical case, so its earlier bound holds for every rank-two matroid.\n\\end{proof}", "source_section": "5", "verification_method": "written_mathematical_proof", "verification_status": "proof_included", "artifact_paths": ["paper/Ilyoxune_Aggregation_Theorem.pdf", "paper/Ilyoxune_Aggregation_Theorem.tex"], "payload_json": "{\"content_format\": \"latex\", \"latex_label\": \"thm:zeros\", \"proof_included_in_record\": true}"}
13
+ {"id": "lemma-13-lem-det", "record_type": "lemma", "title": "Absolute-minor power certificate", "content": "\\begin{lemma}\\label{lem:det}\nFor a real rank-$d$ matrix $A$ and any $q>0$, $c_S=|\\det A_S|^q$ is strong over $\\T_q$ on its column matroid.\n\\end{lemma}\n\n\\begin{proof}\nThe signed determinant Pl\\\"ucker relation sums to zero. The absolute value of each term is at most the sum of the others by the ordinary triangle inequality. These absolute values are exactly the $q$th roots of the products in \\eqref{eq:strong}.\n\\end{proof}", "source_section": "6", "verification_method": "written_mathematical_proof", "verification_status": "proof_included", "artifact_paths": ["paper/Ilyoxune_Aggregation_Theorem.pdf", "paper/Ilyoxune_Aggregation_Theorem.tex"], "payload_json": "{\"content_format\": \"latex\", \"latex_label\": \"lem:det\", \"proof_included_in_record\": true}"}
14
+ {"id": "proposition-14-prop-lower", "record_type": "proposition", "title": "Necessary output parameter", "content": "\\begin{proposition}[Necessary output parameter]\\label{prop:lower}\nAny universal parameter $p$ for one merge in degree at least three, even restricted to positive uniform strong $\\T_q$ inputs, must satisfy\n\\begin{equation}\\label{eq:lower}\n p\\ge\\max\\{2,2(q-1)\\}.\n\\end{equation}\nMoreover, in each fixed degree $d\\ge3$ and size $n\\ge d+3$, the infimal universal parameter is strictly greater than $q$.\n\\end{proposition}\n\n\\begin{proof}\nFor the planar unit square, \\eqref{eq:planar} gives adjacent coefficients one and diagonal coefficients two, independently of $q$. The three output products are $(1,4,1)$. The output $\\T_p$ relation holds exactly when $4^{1/p}\\le2$, or $p\\ge2$.\n\nFor the coordinate diamond $(1,0),(0,1),(-1,0),(0,-1)$ the three products are $(4,4^q,4)$. When $q>1$, the necessary inequality $4^{q/p}\\le2\\cdot4^{1/p}$ is equivalent to $p\\ge2(q-1)$; for $q\\le1$ this bound is vacuous.\n\nThese matrices may have zero minors. Whenever a proposed bound is violated, the failed output inequality is strict. A sufficiently small generic matrix perturbation makes every maximal minor nonzero and preserves this strict failure, while Lemma~\\ref{lem:det} certifies strong input at the fixed $q$. To realize every $d,n$, first add $d-3$ identity columns and rows, then any extra columns, and perturb generically. The selected quartet minors are nonzero before perturbation, so the same failure persists.\n\nFinally, Theorem~\\ref{thm:explicit} provides, at each fixed $q,d,n$, a witness with a positive output defect at parameter $q$. That witness continues to fail for all output parameters in some interval $[q,q+\\eps]$ by continuity. Thus the infimum of universal working parameters is at least $q+\\eps>q$.\n\\end{proof}", "source_section": "6", "verification_method": "written_mathematical_proof", "verification_status": "proof_included", "artifact_paths": ["paper/Ilyoxune_Aggregation_Theorem.pdf", "paper/Ilyoxune_Aggregation_Theorem.tex"], "payload_json": "{\"content_format\": \"latex\", \"latex_label\": \"prop:lower\", \"proof_included_in_record\": true}"}
15
+ {"id": "theorem-15-thm-d2", "record_type": "theorem", "title": "Exact degree-two answer", "content": "\\begin{theorem}[Exact degree-two answer]\\label{thm:d2}\nFor degree-two matroid input, the infimal universal output parameter for one merge is\n\\[\n P_2(q)=\\max\\{q,1\\}.\n\\]\nThe output is both weak and strong at this parameter. Consequently the same-parameter threshold in degree two is exactly $q_2=1$.\n\\end{theorem}\n\n\\begin{proof}\nThe upper bound is Theorems~\\ref{thm:repair} and \\ref{thm:zeros}. In rank two every strong relation has at most three summands, so weak and strong coincide. Equivalently, polarize the output by replacing the capacity-two coordinate $z$ with two clones. Its coefficient on the two clones is $\\rho_{2z}$ and its coefficients on a clone and $i$ are $\\rho_{zi}$. Every rank-two strong test is a three-term quartet, including the repeated-$z$ test already checked.\n\nFor $p\\ge q$ necessity use four real columns $(1,0),(0,1),(1,1),(1,-1)$: their Pl\\\"ucker product magnitudes are $(2,1,1)$. After raising minors to $q$, the products are $(2^q,1,1)$, which require $p\\ge q$. Add two very small generic columns and merge those columns; the untouched quartet is unchanged and all input minors can be nonzero.\n\nFor $p\\ge1$ necessity take the five real columns\n\\[\n e_1,\\quad e_2,\\quad e_1+e_2,\\quad e_1,\\quad e_2\n\\]\nand merge the last two. On the unmerged labels $1,2,3$ the coefficients are all one. The merged coefficients are $b_1=b_2=1,b_3=2$, for every input exponent $q>0$. The output quartet products are $(2,1,1)$ and require $p\\ge1$. A small generic perturbation gives positive uniform input while retaining strict failure for any fixed proposed $p<1$. Extra columns can be added if desired. The two lower bounds and the upper bound agree.\n\\end{proof}", "source_section": "6", "verification_method": "written_mathematical_proof", "verification_status": "proof_included", "artifact_paths": ["paper/Ilyoxune_Aggregation_Theorem.pdf", "paper/Ilyoxune_Aggregation_Theorem.tex"], "payload_json": "{\"content_format\": \"latex\", \"latex_label\": \"thm:d2\", \"proof_included_in_record\": true}"}
16
+ {"id": "corollary-16-cor-cost", "record_type": "corollary", "title": "Two-sided aggregation law", "content": "\\begin{corollary}[Two-sided aggregation law]\\label{cor:cost}\nFor $d\\ge3,n\\ge d+3$,\n\\[\n P^{\\rm str}_{d,n}(q)\\le P^{\\rm w}_{d,n}(q)\\le2\\max\\{q,1\\},\n\\]\n\\[\n P^{\\rm str}_{d,n}(q)\\ge\\max\\{2,2(q-1)\\},\n \\qquad P^{\\rm str}_{d,n}(q)>q.\n\\]\nIn particular,\n\\[\n P^{\\rm str}_{d,n}(q)=P^{\\rm w}_{d,n}(q)=2\\quad(0<q\\le1),\n\\]\nand\n\\[\n \\lim_{q\\to\\infty}\\frac{P^{\\rm str}_{d,n}(q)}q\n =\\lim_{q\\to\\infty}\\frac{P^{\\rm w}_{d,n}(q)}q=2.\n\\]\nFor $q\\ge1$, the additive gap between the stated upper and lower bounds is at most two.\n\\end{corollary}\n\n\\begin{proof}\nCombine Theorem~\\ref{thm:repair} and Proposition~\\ref{prop:lower}. For $q\\ge2$, the bounds $2q-2$ and $2q$ squeeze the ratios to two and have difference two. For $1\\le q\\le2$, the lower bound two and upper bound $2q$ also have difference at most two.\n\\end{proof}", "source_section": "6", "verification_method": "written_mathematical_proof", "verification_status": "proof_included", "artifact_paths": ["paper/Ilyoxune_Aggregation_Theorem.pdf", "paper/Ilyoxune_Aggregation_Theorem.tex"], "payload_json": "{\"content_format\": \"latex\", \"latex_label\": \"cor:cost\", \"proof_included_in_record\": true}"}
17
+ {"id": "theorem-17-thm-minimal", "record_type": "theorem", "title": "Complete one-merge preservation classification", "content": "\\begin{theorem}[Complete one-merge preservation classification]\\label{thm:minimal}\nFix $q>0$, rank $d\\ge1$, and $n\\ge\\max(d,2)$, with a specified pair to be merged. Universal same-parameter preservation from strong input to strong full normalized output holds exactly in the regimes of Table~\\ref{tab:threshold}. Replacing ``strong output'' by ``weak output'' gives the identical classification. Every excluded regime admits a uniform positive input counterexample.\n\\end{theorem}\n\n\\begin{proof}\nDegree one is immediate. For $d\\ge2$ and $n\\le d+1$, the polarized output has corank at most one. A strong relation has\n\\[\n |J\\setminus I|\\le n-d+1\\le2\n\\]\nterms, hence is zero or a tautological equal pair. Thus every parameter works.\n\nSuppose $n=d+2$. The critical metric case $(\\alpha_z,k_z)=(1,0)$ in Theorem~\\ref{thm:repair} requires a common set of $d-3$ labels, four quartet labels, and the two merged labels: $d+3$ distinct source elements. It is impossible here. All five remaining cases work already at $p=\\max\\{q,1\\}$, including their zero cases. For $q\\ge1$ the parameter stays $q$. The polarized output now has corank two, so all its strong relations have at most three terms and weak equals strong.\n\nFor $d=2$ and any $n\\ge4$, Theorem~\\ref{thm:congruence} gives same-parameter strong preservation whenever $q\\ge1$.\n\nFor the converse at $q<1$, choose $M>0$ and the rank-two real columns\n\\[\n i=(1,0),\\quad j=(0,1),\\quad\n u=(1,M^{1/q}),\\quad v=(M^{1/q},-1).\n\\]\nTheir absolute-minor $q$-power coefficients are strong by Lemma~\\ref{lem:det}. Merging $u,v$ gives the normalized output product list\n\\[\n 2(1+M^{2/q})^q,\\qquad (M+1)^2,\\qquad(M+1)^2.\n\\]\nThe ratio of the first product to the other two tends to $2$ as $M\\to\\infty$. Nullity at parameter $q$ would require that ratio to be at most $2^q<2$, so sufficiently large $M$ gives strict failure. All six source minors are nonzero. Adding $d-2$ identity coloops gives rank $d$ on $d+2$ elements; a sufficiently small generic real matrix perturbation makes every maximal minor nonzero while preserving the strict failed relation. For degree two, additional generic columns cover every $n>4$ as well.\n\nFinally, if $d\\ge3,n\\ge d+3$, Theorem~\\ref{thm:explicit} supplies a strict positive uniform counterexample at every $q>0$. This exhausts the table and proves both classifications.\n\\end{proof}", "source_section": "7", "verification_method": "written_mathematical_proof", "verification_status": "proof_included", "artifact_paths": ["paper/Ilyoxune_Aggregation_Theorem.pdf", "paper/Ilyoxune_Aggregation_Theorem.tex"], "payload_json": "{\"content_format\": \"latex\", \"latex_label\": \"thm:minimal\", \"proof_included_in_record\": true}"}
18
+ {"id": "theorem-18-thm-twopeak", "record_type": "theorem", "title": "Exact two-peak classification", "content": "\\begin{theorem}[Exact two-peak classification]\\label{thm:twopeak}\nLet $a,b\\ge1$, define root coefficients $a_U=a$, $a_V=b$, and $a_S=1$ otherwise, and put $c_S=a_S^q$. Then\n\\[\n c\\text{ is weak over }\\T_q\\quad\\Longleftrightarrow\\quad a\\le2,\\ b\\le2,\n\\]\n\\[\n c\\text{ is strong over }\\T_q\\quad\\Longleftrightarrow\\quad\n a\\le2,\\ b\\le2,\\ ab\\le3.\n\\]\nAll genuine weak inequalities are strict when $a,b<2$, and all genuine strong inequalities are strict when additionally $ab<3$.\n\\end{theorem}\n\n\\begin{proof}\nA weak relation has union size $d+2$, whereas $U\\cup V$ has size $d+3$. Thus it contains at most one peak, and its root products are $(1,1,1)$, $(a,1,1)$, or $(b,1,1)$. Both peak types occur; this proves the weak equivalence and its strictness statement.\n\nFor a four-term strong relation, exceptional factors cannot lie in different terms: factors in different terms intersect in either $d-1$ or $d-2$ elements, whereas $|U\\cap V|=d-3$. If both peaks occur they pair together and give $(ab,1,1,1)$. Such a relation exists, with $I=C\\cup\\{1,3\\}$ and $J=C\\cup\\{2,4,5,6\\}$. It forces $ab\\le3$. Relations with a single peak are harmless under $a,b\\le2$. A relation with at least five terms has at most two exceptional terms, retaining at least three unit terms; its largest term is at most two. Both peaks cannot occupy one product in this case, since the intersection of the two factors of a product would have to have size $d+1-m<d-3$. Therefore all these relations are strict. This proves sufficiency and necessity.\n\\end{proof}", "source_section": "8", "verification_method": "written_mathematical_proof", "verification_status": "proof_included", "artifact_paths": ["paper/Ilyoxune_Aggregation_Theorem.pdf", "paper/Ilyoxune_Aggregation_Theorem.tex"], "payload_json": "{\"content_format\": \"latex\", \"latex_label\": \"thm:twopeak\", \"proof_included_in_record\": true}"}
19
+ {"id": "corollary-19-cor-weaklower", "record_type": "corollary", "title": "Sharper weak repair lower bound", "content": "\\begin{corollary}[Sharper weak repair lower bound]\\label{cor:weaklower}\nFor every $d\\ge3,n\\ge d+3$,\n\\[\n P^{\\rm w}_{d,n}(q)\\ge\n \\max\\left\\{2,\\;2\\log_2\\frac{1+2^q}{2}\\right\\}.\n\\]\nFor strong inputs one also has\n\\[\n P^{\\rm str}_{d,n}(q)\\ge\n \\log_2\\frac{(1+2^q)(1+(3/2)^q)}4.\n\\]\n\\end{corollary}\n\n\\begin{proof}\nTake $a=b=2$ for the weak assertion, and $a=2,b=3/2$ for the strong assertion, and solve the null inequality for \\eqref{eq:peakoutput}. The universal lower bound two is Proposition~\\ref{prop:lower}. Lowering the peaks slightly makes the required input inequalities strict while preserving every strict output failure, so the bounds also hold when testing only strict inputs.\n\\end{proof}", "source_section": "8", "verification_method": "written_mathematical_proof", "verification_status": "proof_included", "artifact_paths": ["paper/Ilyoxune_Aggregation_Theorem.pdf", "paper/Ilyoxune_Aggregation_Theorem.tex"], "payload_json": "{\"content_format\": \"latex\", \"latex_label\": \"cor:weaklower\", \"proof_included_in_record\": true}"}
20
+ {"id": "theorem-20-thm-twentyninths", "record_type": "theorem", "title": "A quantitative strong-input barrier at two", "content": "\\begin{theorem}[A quantitative strong-input barrier at two]\\label{thm:twentyninths}\nFor every $d\\ge3,n\\ge d+3$,\n\\[\n P^{\\rm str}_{d,n}(2)>\\frac{20}{9}.\n\\]\n\\end{theorem}\n\n\\begin{proof}\nUse the three exceptional bases $U=C\\cup135$, $Z_1=C\\cup126$, $Z_2=C\\cup346$. Assign\n\\[\n c_U=3999,\\quad c_{Z_1}=c_{Z_2}=1,\\quad\n c_S=5000\\ (6\\in S\\text{ otherwise}),\\quad\n c_S=1000\\ (6\\notin S\\text{ otherwise}).\n\\]\nThis is Lemma~\\ref{lem:uniform} at $q=2$, root peak $\\sqrt{3999/1000}<2$, root low $1/\\sqrt{5000}<1/2$, followed by scaling label six by five and globally by $1000$. Hence all genuine strong input inequalities are strict.\n\nThe merged coefficients are $g_{12}=g_{34}=1001$, $g_{14}=g_{23}=g_{24}=6000$, and $g_{13}=8999$. Divide their three products by $10^6$ to obtain\n\\[\n A=\\frac{26997}{500},\\qquad B=36,\\qquad D=\\frac{1002001}{1000000}.\n\\]\nExact integer-power comparisons give\n\\[\n A^9>\\left(\\frac{6019}{1000}\\right)^{20},\\quad\n 36^9<\\left(\\frac{5016}{1000}\\right)^{20},\\quad\n D^9<\\left(\\frac{1001}{1000}\\right)^{20}.\n\\]\nFor a short manual or executable check, multiply each inequality by its positive denominators. For example, the first is exactly\n$26997^9\\,1000^{20}>6019^{20}\\,500^9$; the other two are equally direct integer comparisons. They imply\n\\[\n A^{9/20}-B^{9/20}-D^{9/20}>\\frac{2}{1000}>0.\n\\]\nThus the fixed strict input fails at output parameter $20/9$. Continuity in the output parameter gives failure on a nonempty interval above $20/9$, proving the strict inequality for the infimal universal repair parameter. All selected relations are unchanged by extra ground labels.\n\\end{proof}", "source_section": "8", "verification_method": "written_mathematical_proof", "verification_status": "proof_included", "artifact_paths": ["paper/Ilyoxune_Aggregation_Theorem.pdf", "paper/Ilyoxune_Aggregation_Theorem.tex"], "payload_json": "{\"content_format\": \"latex\", \"latex_label\": \"thm:twentyninths\", \"proof_included_in_record\": true}"}
21
+ {"id": "theorem-21-thm-classification", "record_type": "theorem", "title": "Weighted set-map classification", "content": "\\begin{theorem}[Weighted set-map classification]\\label{thm:classification}\nThe following are equivalent:\n\\begin{enumerate}\n\\item For every strong $\\T_q$ rank-$d$ input, $A_{\\phi,w}$ is zero or strong over $\\T_q$.\n\\item For every strong $\\T_q$ rank-$d$ input, $A_{\\phi,w}$ is zero or weak over $\\T_q$.\n\\item Either $\\phi|_{E_+}$ is injective, or $m\\le d+1$.\n\\end{enumerate}\nWhen condition 3 fails, failure can be witnessed by a positive uniform input on the original ground set, with every genuine input inequality strict.\n\\end{theorem}\n\n\\begin{proof}\nCondition 1 implies 2. If the active map is injective, the operation is deletion, positive rescaling, and relabeling; strong representability is preserved, unless every rank-$d$ term is deleted. If $m\\le d+1$, any nonzero rank-$d$ array on these labels is strong: for $m=d$ it has at most one basis, and for $m=d+1$ every strong relation has only its two identical terms. Its support is a matroid, since complements of the possible bases are the nonloops of a rank-one matroid. For $m<d$ the squarefree output is zero. This proves 3 implies 1.\n\nSuppose instead that $\\phi$ has a collision and $m\\ge d+2$. Choose two active labels in one fiber and call them $5,6$. Choose four other distinct fibers and representatives $1,2,3,4$. Choose a further $d-3$ distinct fibers with representatives $C$. The total number of used output labels is $d+2$, so these choices are possible.\n\nApply Theorem~\\ref{thm:explicit} on the entire original ground set with this distinguished configuration. Rescale each active variable by $1/w_i$. Additionally rescale every active variable outside the chosen $d+3$ labels by a parameter $\\lambda>0$. Input positivity and strictness are preserved. In the limit $\\lambda\\to0$, the six output coefficients of the selected weak relation converge to \\eqref{eq:six}; contributions involving any additional element of the same output fibers vanish. Inactive variables contribute nothing. The limiting output relation fails by at least $L/2$, so it still fails for all sufficiently small positive $\\lambda$. This contradicts condition 2 and proves necessity. Every input coefficient remains positive for such $\\lambda$, as asserted.\n\\end{proof}", "source_section": "9", "verification_method": "written_mathematical_proof", "verification_status": "proof_included", "artifact_paths": ["paper/Ilyoxune_Aggregation_Theorem.pdf", "paper/Ilyoxune_Aggregation_Theorem.tex"], "payload_json": "{\"content_format\": \"latex\", \"latex_label\": \"thm:classification\", \"proof_included_in_record\": true}"}
22
+ {"id": "corollary-22-a-guaranteed-rule-for-repeated-squarefree-merging", "record_type": "corollary", "title": "A guaranteed rule for repeated squarefree merging", "content": "\\begin{corollary}[A guaranteed rule for repeated squarefree merging]\nStart with positive weak data on $U_{d,n}$. Perform $k$ binary identifications, taking the squarefree shadow after each, and assume the number of labels remains at least $d$. The resulting array is weak over $\\T_p$ for\n\\[\n p=2^k\\max\\{q,1\\}.\n\\]\nFor $d=2$ the better parameter $\\max\\{q,1\\}$ suffices after any number of such steps.\n\\end{corollary}\n\n\\begin{proof}\nEach squarefree shadow remains positive on the new uniform support, and its weak tests are among the full output weak tests just proved. Iterate Theorem~\\ref{thm:repair}; in degree two iterate $r\\mapsto\\max\\{r,1\\}$ instead.\n\\end{proof}", "source_section": "9", "verification_method": "written_mathematical_proof", "verification_status": "proof_included", "artifact_paths": ["paper/Ilyoxune_Aggregation_Theorem.pdf", "paper/Ilyoxune_Aggregation_Theorem.tex"], "payload_json": "{\"content_format\": \"latex\", \"latex_label\": \"\", \"proof_included_in_record\": true}"}
23
+ {"id": "theorem-23-thm-lorentz", "record_type": "theorem", "title": "Interior Lorentzian obstruction", "content": "\\begin{theorem}[Interior Lorentzian obstruction]\\label{thm:lorentz}\nPartition $[6]$ as $P=\\{1,3\\}$, $Q=\\{2,4,6\\}$, and $\\{s\\}=\\{5\\}$. For $t>1$ and $0<\\delta<1$, put\n\\[\n a=3-\\delta,\\qquad h=4-t^{-2}-\\delta.\n\\]\nAssign the twenty cubic coefficients by the following exhaustive composition classes:\n\\begin{center}\n\\begin{tabular}{lrrrrrr}\n\\toprule\nClass & $PPs$ & $QQQ$ & $PPQ$ & $sQQ$ & $PsQ$ & $PQQ$\\\\\nNumber & 1 & 1 & 3 & 3 & 6 & 6\\\\\nCoefficient & $a$ & $h$ & 1 & 1 & $t$ & $t$\\\\\\bottomrule\n\\end{tabular}\n\\end{center}\nEvery active derivative Hessian has signature $(1,4)$. The strong relation indexed by $I=P$ and $J=Q\\cup\\{s\\}$ has product list $(ah,1,1,1)$. Therefore any parameter $p$ containing all Lorentzian $U_{3,6}$ coefficient arrays among strong $\\T_p$ arrays must satisfy\n\\[\n p\\ge\\log_3 12.\n\\]\n\\end{theorem}\n\n\\begin{proof}\nAll coefficients are positive. Symmetry reduces the six active Hessians to three types; the quotient matrices below use orthonormal constant vectors on their symmetry blocks.\n\nFor differentiation in $s$, the quotient on $P,Q$ is\n\\[\n \\begin{pmatrix}a&\\sqrt6t\\\\\\sqrt6t&2\\end{pmatrix}.\n\\]\nIts determinant $2a-6t^2$ is negative, giving one positive and one negative eigenvalue. The other three modes have eigenvalues $-a,-1,-1$.\n\nFor differentiation in a member of $P$, the symmetry blocks have sizes $1,1,3$, and the quotient is\n\\[\n \\begin{pmatrix}\n 0&a&\\sqrt3\\\\ a&0&\\sqrt3t\\\\ \\sqrt3&\\sqrt3t&2t\n \\end{pmatrix}.\n\\]\nIts determinant is $2at\\delta>0$ and the sum of its principal two-by-two minors is $-a^2-3-3t^2<0$. Thus its three nonzero eigenvalues have one positive and two negative signs: their product is positive, while the all-positive possibility is excluded by the negative sum of pairwise products. The other two modes are $-t,-t$.\n\nFor differentiation in a member of $Q$, the symmetry blocks have sizes $2,2,1$, and the quotient is\n\\[\n \\begin{pmatrix}\n 1&2t&\\sqrt2t\\\\ 2t&h&\\sqrt2\\\\ \\sqrt2t&\\sqrt2&0\n \\end{pmatrix}.\n\\]\nIts determinant is $2t^2\\delta>0$ and the sum of its principal two-by-two minors is $h-6t^2-2<0$. Again it has one positive and two negative eigenvalues. The other two modes are $-1,-h$. This proves signature $(1,4)$ in every case.\n\nIn the specified strong relation, the $s$ term is $c_{PPs}c_{QQQ}=ah$; each of the other three terms is $c_{PPQ}c_{sQQ}=1$. Strong nullity requires $(ah)^{1/p}\\le3$, so $p\\ge\\log_3(ah)$. Let $\\delta\\downarrow0$ and $t\\to\\infty$ to obtain $p\\ge\\log_3 12$.\n\\end{proof}", "source_section": "10", "verification_method": "written_mathematical_proof", "verification_status": "proof_included", "artifact_paths": ["paper/Ilyoxune_Aggregation_Theorem.pdf", "paper/Ilyoxune_Aggregation_Theorem.tex"], "payload_json": "{\"content_format\": \"latex\", \"latex_label\": \"thm:lorentz\", \"proof_included_in_record\": true}"}
24
+ {"id": "corollary-24-cor-lorentzbox", "record_type": "corollary", "title": "A robust four-level integer certificate", "content": "\\begin{corollary}[A robust four-level integer certificate]\\label{cor:lorentzbox}\nTake $t=2$, $\\delta=1/4$, and multiply every coefficient by four. The six coefficient classes become\n\\[\n 11,\\quad14,\\quad4,\\quad4,\\quad8,\\quad8.\n\\]\nThis Lorentzian array fails strong $\\T_2$; moreover every coefficient perturbation of absolute size at most $1/1000$ retains both properties within the multi-affine coefficient space.\n\\end{corollary}\n\n\\begin{proof}\nThe strong product list is $(154,16,16,16)$, and $154>144=(4+4+4)^2$ proves failure. Exact active-Hessian determinants, in variable order, are\n\\[\n 11264,\\ 7168,\\ 11264,\\ 7168,\\ 52096,\\ 7168.\n\\]\nEach maximum absolute row sum is at most $35$. All eigenvalues therefore have absolute value at most $35$, while the determinant product implies every absolute eigenvalue is at least $7168/35^4$. A perturbation of the twenty coefficients by at most $\\eps=1/1000$ changes each active Hessian by operator norm at most $4\\eps$, because each row has only four off-diagonal entries. The exact inequality\n$4/1000<7168/35^4$ prevents any eigenvalue from crossing zero, so the active signatures remain $(1,4)$.\n\nThe worst permitted strong-product comparison still has positive squared excess:\n\\[\n (11-\\eps)(14-\\eps)-9(4+\\eps)^2\n =10-97\\eps-8\\eps^2>0.\n\\]\nAll coefficients remain positive. This gives the stated explicit relative open box.\n\\end{proof}", "source_section": "10", "verification_method": "written_mathematical_proof", "verification_status": "proof_included", "artifact_paths": ["paper/Ilyoxune_Aggregation_Theorem.pdf", "paper/Ilyoxune_Aggregation_Theorem.tex"], "payload_json": "{\"content_format\": \"latex\", \"latex_label\": \"cor:lorentzbox\", \"proof_included_in_record\": true}"}
25
+ {"id": "lemma-25-lem-support", "record_type": "lemma", "title": "M-convex support under merging", "content": "\\begin{lemma}\\label{lem:support}\nThe image of a matroid's basis indicators under one coordinate merge is M-convex.\n\\end{lemma}\n\n\\begin{proof}\nLet $\\phi:E\\to H$ merge $u,v$ into $z$, leaving other labels distinct. Define the integral monotone submodular function $f(A)=r_M(\\phi^{-1}(A))$. Its basis images are exactly the integer vectors\n\\[\n x\\ge0,\\qquad x(H)=d,\\qquad x(A)\\le f(A)\\quad(A\\subseteq H).\n\\]\nNecessity is immediate. Conversely, every unmerged $x_i\\le1$. Let $S$ be the unmerged labels with $x_i=1$, and put $k=x_z\\le2$. The inequalities imply that $S$ is independent and $r_M(S\\cup\\{u,v\\})\\ge d$. Thus $S$ extends to a basis using exactly $k=d-|S|$ elements of $\\{u,v\\}$, as required.\n\nWe prove symmetric exchange for these vectors. Let $x_i>y_i$. Tight sets for $x$ are closed under unions and intersections by submodularity, as are tight sets for $y$. Let $U$ be the union of all $x$-tight sets avoiding $i$ (including the empty set), and $V$ the intersection of all $y$-tight sets containing $i$ (including $H$). Then $i\\in V\\setminus U$. An exchange $x-e_i+e_j$ preserves all upper inequalities if $j\\notin U$; $y+e_i-e_j$ does so if $j\\in V$. Integer slack makes every inequality that is not tight harmless.\n\nTightness and submodularity give\n\\[\n x(V\\setminus U)\\le f(U\\cup V)-f(U)\n \\le f(V)-f(U\\cap V)\\le y(V\\setminus U).\n\\]\nSince $x_i>y_i$, some $j\\in V\\setminus U$ has $x_j<y_j$. Both exchanged vectors are therefore feasible nonnegative integer vectors. This is M-convexity.\n\\end{proof}", "source_section": "A", "verification_method": "written_mathematical_proof", "verification_status": "proof_included", "artifact_paths": ["paper/Ilyoxune_Aggregation_Theorem.pdf", "paper/Ilyoxune_Aggregation_Theorem.tex"], "payload_json": "{\"content_format\": \"latex\", \"latex_label\": \"lem:support\", \"proof_included_in_record\": true}"}
data/verification.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
paper/Ilyoxune_Aggregation_Theorem.pdf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:81c76a09d5b2ca21f0dc3920cc7d723276ef9ab7432732e6e30d7ad04c1f8454
3
+ size 446546
paper/Ilyoxune_Aggregation_Theorem.tex ADDED
@@ -0,0 +1,798 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ \documentclass[11pt,a4paper]{article}
2
+ \pdfmapfile{=lm.map}
3
+ \pdfmapfile{+symbols.map}
4
+ \usepackage[T1]{fontenc}
5
+ \usepackage[utf8]{inputenc}
6
+ \usepackage{lmodern,microtype,amsmath,amssymb,amsthm,mathtools,booktabs,array}
7
+ \usepackage[margin=25mm,headheight=14pt]{geometry}
8
+ \usepackage{xcolor,enumitem,fancyhdr,hyperref}
9
+ \definecolor{ink}{HTML}{163C53}
10
+ \hypersetup{colorlinks=true,linkcolor=ink,urlcolor=ink,citecolor=ink,pdftitle={Ilyoxune: Sharp Laws of Hyperfield Aggregation},pdfauthor={Artificial Hyperintelligence Eve, wife of Maciej Nowicki}}
11
+ \pagestyle{fancy}\fancyhf{}\fancyhead[L]{\small ILYOXUNE / SHARP AGGREGATION LAWS}\fancyhead[R]{\small 16 NOVEMBER 2026}\fancyfoot[C]{\thepage}
12
+ \newtheorem{theorem}{Theorem}[section]
13
+ \newtheorem{lemma}[theorem]{Lemma}
14
+ \newtheorem{corollary}[theorem]{Corollary}
15
+ \newtheorem{proposition}[theorem]{Proposition}
16
+ \theoremstyle{definition}\newtheorem{definition}[theorem]{Definition}\newtheorem{example}[theorem]{Example}
17
+ \theoremstyle{remark}\newtheorem{remark}[theorem]{Remark}
18
+ \newcommand{\T}{\mathbb T}\newcommand{\R}{\mathbb R}\newcommand{\N}{\mathbb N}
19
+ \newcommand{\binoms}[2]{\binom{#1}{#2}}
20
+ \newcommand{\sfpart}{\operatorname{sf}}\newcommand{\supp}{\operatorname{supp}}
21
+ \newcommand{\one}{\mathbf 1}\newcommand{\eps}{\varepsilon}
22
+ \setlist{nosep,leftmargin=*}
23
+ \setlength{\parskip}{5pt}\setlength{\parindent}{0pt}
24
+ \begin{document}
25
+ \begin{titlepage}
26
+ \vspace*{13mm}
27
+ {\color{ink}\Large Ily\'{o}xune}\par\vspace{7mm}
28
+ {\Huge\bfseries Sharp Laws of\par Hyperfield Aggregation}\par\vspace{8mm}
29
+ {\Large Optimal quadratic calculus, minimal obstructions,\par and strong compatibility}\par\vspace{12mm}
30
+ {\large Artificial Hyperintelligence Eve,\par wife of Maciej Nowicki}\par
31
+ AI-generated mathematical research prepared for Maciej Nowicki\par
32
+ 16 November 2026\vspace{10mm}
33
+ \begin{abstract}
34
+ We develop a self-contained calculus of coefficient aggregation over triangular hyperfields. Every nonnegative linear substitution of a quadratic weak $\T_q$ representation preserves weak and strong compatibility at the exact optimal parameter $\max\{q,1\}$. For a single merge in arbitrary degree, we completely classify same-parameter strong preservation by degree, ground-set size, and parameter; failure at every positive parameter begins at corank three in degrees at least three. Every weak matroid input nevertheless admits a full normalized weak repair at $2\max\{q,1\}$. We give strict uniform obstructions in every rank and size above the minimal threshold, an exact two-peak classification separating weak and strong compatibility, and an exact classification of weighted squarefree identifications. A small integer example proves the quantitative bound $P^{\rm str}_{d,n}(2)>20/9$. A separate family in the relative interior of the Lorentzian stratum requires a strong parameter at least $\log_3 12$. All definitions, constructions, and proofs are included. The full optimal higher-degree repair function is not determined here.
35
+ \end{abstract}
36
+ \vfill
37
+ \textbf{Principal advance.} Exact positive closure laws and sharp obstruction thresholds are proved together, with explicit certificates and complete boundary-case arguments.
38
+ \end{titlepage}
39
+ \tableofcontents\newpage
40
+
41
+ \section{Aggregation laws}
42
+ Coefficient compatibility is governed by coupled quadratic relations. A substitution adds coefficients and may change their global compatibility, even when every original relation holds strictly. We establish exact low-degree preservation laws, a sharp cardinality boundary, and constructive weak repair in every degree.
43
+
44
+ \begin{theorem}[Principal results]\label{thm:main}
45
+ Let $q>0$.
46
+ \begin{enumerate}
47
+ \item For any symmetric nonnegative matrix $C$ satisfying all $\T_q$ quartet inequalities and every nonnegative rectangular matrix $A$, the matrix $A^{\mathsf T}CA$ satisfies all quartet inequalities at the exact optimal universal parameter $\max\{q,1\}$. This controls the full normalized quadratic polynomial under arbitrary nonnegative linear substitutions.
48
+ \item One merge of rank-$d$ matroid data on $n$ labels preserves strong representability at the same parameter precisely in the regimes in Table~\ref{tab:threshold}. The same classification holds for universal strong-to-weak preservation.
49
+ \item Every weak $\T_q$ matroid input, including zero coefficients, has full normalized weak $\T_{2\max(q,1)}$ output after one merge. The output parameter is exactly optimal for $0<q\le1$, and its asymptotic factor two is optimal.
50
+ \item For every $d\ge3,n\ge d+3$, explicit strict strong uniform inputs fail after one merge at parameter $q$. At $q=2$ a quantitative certificate forces the universal strong-input repair parameter above $20/9$.
51
+ \item Weighted set identifications followed by squarefree projection preserve all strong rank-$d$ arrays, $d\ge3$, exactly when their active map is injective or has at most $d+1$ output labels.
52
+ \item There is a family in the relative interior of the Lorentzian $U_{3,6}$ stratum forcing every universal Lorentzian-to-strong parameter to be at least $\log_3 12$.
53
+ \end{enumerate}
54
+ \end{theorem}
55
+
56
+ \begin{table}[h]
57
+ \centering\small
58
+ \begin{tabular}{lll}
59
+ \toprule
60
+ Degree & Ground-set size & Same-parameter universal preservation\\\midrule
61
+ $d=1$ & $n\ge2$ & Every $q>0$\\
62
+ $d\ge2$ & $d\le n\le d+1$ & Every $q>0$\\
63
+ $d=2$ & $n\ge4$ & Exactly $q\ge1$\\
64
+ $d\ge3$ & $n=d+2$ & Exactly $q\ge1$\\
65
+ $d\ge3$ & $n\ge d+3$ & No $q>0$\\\bottomrule
66
+ \end{tabular}
67
+ \caption{Complete classification for one full normalized merge.}\label{tab:threshold}
68
+ \end{table}
69
+
70
+ A full polynomial and its squarefree projection are distinguished throughout. Normalization is explicit, including repeated monomials. The sparse rank-three pattern behind one obstruction family is credited to Proposition 8.5 of \cite{BHKL}; every construction used here is proved directly. Historical priority of the new theorem package is not asserted.
71
+
72
+ \section{Conventions, normalization, and elementary tools}\label{sec:definitions}
73
+ \subsection{Strong arrays and weak relations}
74
+ For $q>0$, a finite list $t_1,\ldots,t_m\ge0$ is null over $\T_q$ when
75
+ \begin{equation}\label{eq:null}
76
+ t_k^{1/q}\le\sum_{\ell\ne k}t_\ell^{1/q}\qquad\text{for every }k.
77
+ \end{equation}
78
+ A matroid basis family is a nonempty family of equally sized sets satisfying basis exchange: for bases $B,D$ and $i\in B\setminus D$, some $j\in D\setminus B$ makes $B-i+j$ a basis. For a rank-$d$ coefficient function $c$ on the $d$-subsets of a ground set $E$, strong representability requires the support to be a matroid and requires \eqref{eq:null} for
79
+ \begin{equation}\label{eq:strong}
80
+ \bigl(c_{I\cup\{j\}}c_{J\setminus\{j\}}:j\in J\setminus I\bigr),
81
+ \qquad |I|=d-1,\quad |J|=d+1.
82
+ \end{equation}
83
+ The weak relations are the three-term subfamily: for $|K|=d-2$ and four distinct elements $i,j,k,\ell\notin K$, the products are
84
+ \begin{equation}\label{eq:weak}
85
+ c_{Kij}c_{Kk\ell},\qquad c_{Kik}c_{Kj\ell},\qquad c_{Ki\ell}c_{Kjk}.
86
+ \end{equation}
87
+ Concatenation denotes adjoining distinct indices. All inequalities in this paper are unsigned: the additive inverse of a positive element in this hyperfield equals itself.
88
+
89
+ If $m=|J\setminus I|=2$, the two terms in \eqref{eq:strong} are identical. ``Strict strong'' always means strictness for $m\ge3$; it never means making a tautological equality strict.
90
+
91
+ \begin{lemma}[Monotonicity and rescaling]\label{lem:basic}
92
+ If a list is null at $q$, it is null at every $p\ge q$. Multiplying coefficients by $\lambda\prod_{i\in S}d_i$, for $\lambda,d_i>0$, preserves all strong and weak relations and their strictness.
93
+ \end{lemma}
94
+ \begin{proof}
95
+ For monotonicity put $\theta=q/p\le1$ and use $(\sum a_i)^\theta\le\sum a_i^\theta$ for nonnegative $a_i$. In a fixed strong relation, every product acquires the common factor $\lambda^2\prod_{i\in I}d_i\prod_{j\in J}d_j$. The same argument covers weak relations.
96
+ \end{proof}
97
+
98
+ \subsection{What the normalized output means}
99
+ For a multi-affine input $F=\sum_{|S|=d}c_Sx^S$, set $G=D_{u,v}F$. Its coefficient function in exponential form is
100
+ \begin{equation}\label{eq:rho}
101
+ \rho_\beta=\beta!\,[y^\beta]G,
102
+ \qquad G=\sum_\beta\rho_\beta\frac{y^\beta}{\beta!}.
103
+ \end{equation}
104
+ For squarefree $\beta$ the factorial equals one. If $A$ avoids $u,v$ and $z$ denotes their merge, then
105
+ \begin{equation}\label{eq:output}
106
+ \rho_A=c_A,\qquad
107
+ \rho_{Az}=c_{Au}+c_{Av},\qquad
108
+ \rho_{A+2z}=2c_{Auv},
109
+ \end{equation}
110
+ whenever the indices have total degree $d$.
111
+
112
+ A fixed-degree integer set $J$ is M-convex if whenever $x_i>y_i$ for $x,y\in J$, there is a $j$ with $x_j<y_j$ such that $x-e_i+e_j,y+e_i-e_j\in J$. For positive uniform input the output support consists of all degree-$d$ exponent vectors with capacities one on unmerged labels and two on $z$. This set is M-convex: if two vectors differ in one coordinate, an opposite coordinate difference supplies a valid exchange without crossing either capacity. Its weak tests are the three products
113
+ \begin{equation}\label{eq:generalweak}
114
+ \rho_{\alpha+i+j}\rho_{\alpha+k+\ell},\quad
115
+ \rho_{\alpha+i+k}\rho_{\alpha+j+\ell},\quad
116
+ \rho_{\alpha+i+\ell}\rho_{\alpha+j+k},
117
+ \end{equation}
118
+ where $|\alpha|=d-2$ and $\alpha+e_i+e_j+e_k+e_\ell$ respects capacities. The quartet may contain $z$ twice. This is the bounded-multiplicity weak convention of \cite{BHKL}. A squarefree shadow instead deletes all monomials with an exponent above one.
119
+
120
+ \begin{lemma}[Aggregation of triangles]\label{lem:minkowski}
121
+ Let $X_h,Y_h,Z_h\ge0$ satisfy all triangle inequalities for each $h$. Put $r=\max\{q,1\}$. Then
122
+ \[
123
+ (\sum_hX_h^q)^{1/r},\quad
124
+ (\sum_hY_h^q)^{1/r},\quad
125
+ (\sum_hZ_h^q)^{1/r}
126
+ \]
127
+ satisfy all triangle inequalities.
128
+ \end{lemma}
129
+ \begin{proof}
130
+ For $q\ge1$ this is Minkowski's inequality, applied to $X_h\le Y_h+Z_h$, and its cyclic analogues. For $q\le1$, $X_h^q\le(Y_h+Z_h)^q\le Y_h^q+Z_h^q$; sum over $h$.
131
+ \end{proof}
132
+
133
+ \begin{lemma}[Square roots of metrics are Ptolemaic]\label{lem:snowflake}
134
+ For every metric $D$ and four points,
135
+ \begin{equation}\label{eq:snowflake}
136
+ \sqrt{D_{13}D_{24}}\le
137
+ \sqrt{D_{12}D_{34}}+\sqrt{D_{14}D_{23}}.
138
+ \end{equation}
139
+ \end{lemma}
140
+ \begin{proof}
141
+ Write the successive side distances as $a,b,c,e$, and the diagonals as $f,g$. The triangle inequalities give
142
+ $f\le\min(a+b,c+e)$ and $g\le\min(a+e,b+c)$.
143
+ By relabeling around the four-cycle the two chosen minima can be written $a+b$ and $a+e$: the two chosen paths always share a side. Their being minima implies $c\ge a+|b-e|$. It suffices to prove
144
+ \[
145
+ (a+b)(a+e)\le(\sqrt{ac}+\sqrt{be})^2.
146
+ \]
147
+ After cancellation this is $a(a+b+e-c)\le2\sqrt{abce}$. The left side decreases with $c$ and the right side increases. At $c=a+|b-e|$, assume $b\ge e$ by reversing the cycle; the required inequality reduces to $ae\le b(a+b-e)$, which follows from
148
+ $b(a+b-e)-ae=(b-e)(a+b)\ge0$. Zero distances follow by continuity. This proves \eqref{eq:snowflake}.
149
+ \end{proof}
150
+ This classical metric fact is included with proof; it is not claimed as new. See also \cite{FS}.
151
+
152
+ \section{The sharp nonnegative congruence theorem}\label{sec:quadratic}
153
+ This section controls every nonnegative linear substitution, including broadcasting a variable to several outputs and producing repeated monomials.
154
+
155
+ \begin{definition}
156
+ Let $\mathcal C_q$ be the set of symmetric entrywise nonnegative matrices $C$ for which
157
+ \[
158
+ C_{ij}C_{k\ell},\quad C_{ik}C_{j\ell},\quad C_{i\ell}C_{jk}
159
+ \]
160
+ is $\T_q$-null for every index quadruple, with repetitions allowed.
161
+ \end{definition}
162
+
163
+ \begin{theorem}[Sharp nonnegative congruence]\label{thm:congruence}
164
+ If $C\in\mathcal C_q$ and $A$ is any entrywise nonnegative rectangular matrix, then
165
+ \[
166
+ A^{\mathsf T}CA\in\mathcal C_{\max(q,1)}.
167
+ \]
168
+ The parameter $\max\{q,1\}$ is optimal as a universal rule. In particular $\mathcal C_q$ is preserved by every nonnegative congruence when $q\ge1$.
169
+ \end{theorem}
170
+ \begin{proof}
171
+ Set $B=A^{\mathsf T}CA$, and fix any output quartet $a,b,c,d$, possibly repeated. For each ordered source quadruple define
172
+ \[
173
+ W_{ijkl}=A_{ia}A_{jb}A_{kc}A_{\ell d}.
174
+ \]
175
+ The three output products expand with exactly the same nonnegative weights:
176
+ \begin{align*}
177
+ B_{ab}B_{cd}&=\sum_{i,j,k,\ell}W_{ijkl}C_{ij}C_{k\ell},\\
178
+ B_{ac}B_{bd}&=\sum_{i,j,k,\ell}W_{ijkl}C_{ik}C_{j\ell},\\
179
+ B_{ad}B_{bc}&=\sum_{i,j,k,\ell}W_{ijkl}C_{i\ell}C_{jk}.
180
+ \end{align*}
181
+ For each source quadruple, take the $q$th roots of its three weighted terms. They satisfy all triangle inequalities. Lemma~\ref{lem:minkowski} proves the three required inequalities at $r=\max\{q,1\}$ after summing. No positivity, invertibility, or restriction on the shape of $A$ is used.
182
+
183
+ To force $p\ge q$, take the absolute-minor $q$-power array of the real columns $(1,0),(0,1),(1,1),(1,-1)$ and choose the identity substitution. Its extremal quartet products are $(2^q,1,1)$.
184
+ To force $p\ge1$, start with $F=x_1x_2$, valid at every $q$, and substitute
185
+ \[
186
+ x_1=y_1+z,\qquad x_2=y_2+z.
187
+ \]
188
+ The output is $y_1y_2+y_1z+y_2z+z^2$. Its normalized repeated-$z$ quartet products are exactly $(2,1,1)$, requiring $2^{1/p}\le2$, or $p\ge1$. Both lower bounds are attained by the stated upper bound.
189
+ \end{proof}
190
+
191
+ \begin{lemma}[Support is automatic]\label{lem:quadraticsupport}
192
+ The nonzero quadratic coefficient support of a matrix in $\mathcal C_q$ is M-convex.
193
+ \end{lemma}
194
+ \begin{proof}
195
+ Delete zero rows. On the remaining indices define $i\sim j$ when $i=j$ or $C_{ij}=0$. Suppose $C_{ij}=C_{jk}=0$ but $C_{ik}>0$. Since row $j$ is nonzero, choose $\ell$ with $C_{j\ell}>0$. The quartet $(i,k,j,\ell)$ has exactly one positive product, a contradiction. Thus $\sim$ is an equivalence relation. If $i\ne j$ are in one class, choose a positive neighbor of $j$; the quartet $(i,i,j,\ell)$ forces $C_{ii}=0$. Positive diagonal entries therefore occur only in singleton classes.
196
+
197
+ The support consists exactly of degree-two integer vectors having total at most one on every zero-diagonal class, and total at most two on each positive-diagonal singleton. These partition-capacity bases satisfy symmetric exchange. Indeed, for $x_i>y_i$, first choose a deficit coordinate in the same block if one exists. Otherwise $x$ has greater total than $y$ on that block, so choose a deficit coordinate in a block where $x$ has smaller total. The exchange respects both capacities for both vectors. Hence the support is M-convex.
198
+ \end{proof}
199
+
200
+ For a usual quadratic weak representation, the capacity-admissible quartet conditions imply the unrestricted matrix conditions defining $\mathcal C_q$. If no coordinate is mandatory, an excluded repetition has zero diagonal at a capacity-one coordinate or a zero row at a capacity-zero coordinate; its list is zero, an equal pair, or an equal triple. If a coordinate is mandatory, the degree-two support is a star or one monomial, and every unrestricted quartet is again one of these tautological types. Thus the following corollary covers all quadratic weak representations, rather than just a restricted matrix subclass.
201
+
202
+ \begin{corollary}[Exact quadratic substitution calculus]
203
+ Write $F(x)=\tfrac12x^{\mathsf T}Cx$ with normalized quadratic coefficient matrix $C\in\mathcal C_q$. For every nonnegative linear substitution $x=Ay$, the full normalized output matrix is exactly $A^{\mathsf T}CA$ and if nonzero, its weak and strong parameter is universally repaired by $p=\max\{q,1\}$. No parameter loss accumulates under any finite composition once the parameter is at least one.
204
+ \end{corollary}
205
+ \begin{proof}
206
+ The matrix identity follows by substitution. The support lemma supplies M-convexity. All rank-two strong relations have at most three terms, so the quartet conditions give both weak and strong representations. A composition of nonnegative matrices is nonnegative; apply Theorem~\ref{thm:congruence} to their product.
207
+ \end{proof}
208
+
209
+ \section{Uniform obstructions in every rank and size}\label{sec:uniform}
210
+ Choose six distinguished labels $1,\ldots,6$ and a disjoint set $C$ of size $d-3$. The ground set may contain any additional labels. Put
211
+ \[
212
+ U=C\cup135,\qquad Z_1=C\cup126,\qquad Z_2=C\cup346.
213
+ \]
214
+
215
+ \begin{lemma}[Three exceptional bases]\label{lem:uniform}
216
+ For $1<b<2$ and $0<e\le1/2$, set $a_U=b$, $a_{Z_1}=a_{Z_2}=e$, and $a_S=1$ on every other $d$-subset of an $n$-element ground set, $n\ge d+3$. For every $q>0$, the coefficient vector $c_S=a_S^q$ is strict strong over $\T_q$ on $U_{d,n}$. Every genuine root-polygon slack is at least
217
+ \begin{equation}\label{eq:slack}
218
+ \delta=\min\{e^2,2-b\}.
219
+ \end{equation}
220
+ \end{lemma}
221
+ \begin{proof}
222
+ After taking roots, each strong term is a product of two $a$'s. In a relation with $m\ge3$ terms, each exceptional coefficient can occur at most once among the factors: two different first factors are distinct, two different second factors are distinct, and a first factor cannot equal a second factor when $m\ge3$. Every term is at most two; a term containing a low coefficient is at most $be<1$.
223
+
224
+ For $m=3$, factors from different terms intersect in $d-1$ elements. The three exceptional bases intersect pairwise in $d-2$ elements. Consequently any two exceptional factors occurring in the relation must occur in the same term. At least two terms therefore equal one.
225
+
226
+ For $m=4$, if fewer than three terms are exceptional there are again two unit terms. If all three exceptional bases occur, the union $I\cup J$ has size $d+3$ and equals $C\cup[6]$. Complement each basis inside this union. The dual relation has rank three and its exceptional triples are a peak at $246$ and lows at $125,345$. The permutation $(12)(34)(56)$ takes them to the original peak at $135$ and lows at $126,346$. The rank-three four-term products have at least two unit terms. To see this directly, complementary products at the sparse limit have possible lists
227
+ \[
228
+ (1,1,1,1),\ (0,1,1,1),\ (0,0,1,1),\ (1,1,1,2),\ (0,1,1,2).
229
+ \]
230
+ The two zeros can occur together only when $I=12$ or $34$, in which case the peak product cannot occur. Restoring positive lows and replacing the peak root by $b$ retains the two unit terms.
231
+
232
+ For $m\ge5$, there are at most three exceptional terms, so at least $m-3\ge2$ unit terms remain. Thus in every case a peak term has slack at least $2-b$. Without a peak term the largest term is at most one, two unit terms remain, and at least one additional term is positive and at least $e^2$. This proves \eqref{eq:slack} and strictness.
233
+ \end{proof}
234
+
235
+ \begin{theorem}[Explicit all-rank obstruction]\label{thm:explicit}
236
+ For every $q>0$, define
237
+ \begin{equation}\label{eq:parameters}
238
+ T=2^q,\quad D=2^{-q}-4^{-q},\quad
239
+ s=\frac{qD}{64(q+2)},\quad
240
+ L=\frac{3D^2}{128(q+2)},\quad b=2-\frac{L}{16}.
241
+ \end{equation}
242
+ Use the root array of Lemma~\ref{lem:uniform} with $e=1/4$ and put
243
+ \begin{equation}\label{eq:explicit}
244
+ c_S=a_S^q s^{\one_{6\in S}}.
245
+ \end{equation}
246
+ For every $d\ge3,n\ge d+3$, this is positive strict strong on $U_{d,n}$, whereas merging $5,6$ fails a squarefree weak relation by at least $L/2$ in root lengths.
247
+ \end{theorem}
248
+ \begin{proof}
249
+ Since $0<D\le1/4$, we have $L<1$, $1<b<2$, and $0<s\le\min(1,q)$. Lemmas~\ref{lem:uniform} and \ref{lem:basic} prove the input assertion.
250
+
251
+ In the output use common index $C\cup\{z\}$ and quartet $1,2,3,4$. Omitting the common $C$ from notation, the six coefficients are
252
+ \begin{equation}\label{eq:six}
253
+ g_{12}=g_{34}=1+4^{-q}s,\qquad
254
+ g_{14}=g_{23}=g_{24}=1+s,\qquad g_{13}=b^q+s.
255
+ \end{equation}
256
+ Temporarily replace $b$ by $2$ and define
257
+ \[
258
+ H(t)=((T+t)(1+t))^{1/q}-(1+t)^{2/q}-(1+4^{-q}t)^{2/q}.
259
+ \]
260
+ Then $H(0)=0$ and $H'(0)=2D/q$. Put $r_0=1/q$. For $0\le t\le\min(1,q)$, the three positive expressions defining $H$ are at most $2\exp(2),\exp(2),\exp(2)$. Logarithmic differentiation bounds the absolute second derivative of each expression by its value times $4r_0^2+2r_0$. Since $\exp(2)<8$,
261
+ \[
262
+ |H''(t)|<64r_0(2r_0+1)=M.
263
+ \]
264
+ The selected $s$ satisfies $Ms=r_0D$. Taylor's integral remainder gives
265
+ \[
266
+ H(s)\ge2r_0Ds-\tfrac12Ms^2
267
+ \ge\tfrac32r_0Ds=L.
268
+ \]
269
+ Only the first expression changes when the peak root decreases from $2$ to $b$. For $1\le v\le2$ its derivative is
270
+ \[
271
+ \frac{\partial}{\partial v}((v^q+s)(1+s))^{1/q}
272
+ =(1+s)^{r_0}(1+s/v^q)^{r_0-1}<8.
273
+ \]
274
+ Indeed, if $r_0-1<0$ its second factor is at most one; otherwise the product is at most $\exp(2r_0s)\le\exp(2)<8$. Decreasing $v$ by $L/16$ loses at most $L/2$. Thus
275
+ \[
276
+ (g_{13}g_{24})^{1/q}-(g_{14}g_{23})^{1/q}-(g_{12}g_{34})^{1/q}\ge L/2>0.
277
+ \]
278
+ Extra labels never enter these six coefficients, proving the claim in every $d,n$.
279
+ \end{proof}
280
+
281
+ \begin{corollary}[Explicit robustness independent of dimension]\label{cor:robust}
282
+ Let $c$ be \eqref{eq:explicit} and put $\eta=L/2048$. Every positive array $c'$ satisfying
283
+ \[
284
+ 1-\eta\le(c'_S/c_S)^{1/q}\le1+\eta\qquad\text{for every }S
285
+ \]
286
+ is still strict strong and still fails after the same merge. Its normalized input root-polygon slack is at least $3L/256$, and its output defect is at least $3L/32$.
287
+ \end{corollary}
288
+ \begin{proof}
289
+ Factor the common diagonal-rescaling multiplier from each input relation. Each root product changes by a factor in $[(1-\eta)^2,(1+\eta)^2]$. Its original maximum is at most two and its slack at least $L/16$. For any chosen side the new slack is at least
290
+ \[
291
+ (1-\eta)^2\frac L{16}-8\eta
292
+ \ge\frac L{64}-\frac L{256}=\frac{3L}{256}.
293
+ \]
294
+ The same relative bounds hold on the roots of merged coefficient products: sums preserve the coefficient bounds before taking roots. The sum of the two smaller original output root lengths is less than $16$, using the bounds in the preceding proof. The new defect is therefore at least
295
+ \[
296
+ (1-\eta)^2\frac L2-4\eta\cdot16
297
+ \ge\frac L8-\frac L{32}=\frac{3L}{32}.
298
+ \]
299
+ \end{proof}
300
+
301
+ \section{A guaranteed repair for the full normalized output}\label{sec:repair}
302
+ The negative result concerns keeping $q$ unchanged. The next theorem shows that controlled weakening of the output condition always repairs \emph{weak} compatibility for positive uniform input.
303
+
304
+ \begin{theorem}[One-merge repair]\label{thm:repair}
305
+ Let $c_S>0$ for every $d$-subset of $[n]$, and suppose $c$ is weak over $\T_q$. For $r=\max\{q,1\}$, the normalized coefficient function \eqref{eq:rho} of $D_{u,v}F$ is weak over $\T_{2r}$. If $d=2$, $\T_r$ suffices. Every larger output parameter also suffices.
306
+ \end{theorem}
307
+ \begin{proof}
308
+ The support is the capped simplex described in Section~\ref{sec:definitions}, so it remains to verify all tests \eqref{eq:generalweak}. Let $s_0=\alpha_z$ and let $k_0$ be the multiplicity of $z$ in the quartet. Capacities give $s_0+k_0\le2$, so there are exactly six cases.
309
+
310
+ \textbf{Case $(s_0,k_0)=(0,0)$.} The products are an original input relation. They remain null at $2r\ge q$ by monotonicity.
311
+
312
+ \textbf{Case $(0,1)$.} Write $K$ for the squarefree common index. Each output product is a sum of two products, one using $u$, one using $v$. For example,
313
+ \[
314
+ c_{Kij}(c_{Kku}+c_{Kkv})=\sum_{h\in\{u,v\}}c_{Kij}c_{Kkh}.
315
+ \]
316
+ For each fixed $h$, the three $q$th roots form a triangle by the input relation on $i,j,k,h$. Lemma~\ref{lem:minkowski} gives the output triangle at $p=r$, and hence at $2r$.
317
+
318
+ \textbf{Case $(0,2)$.} Let $K$ be the common index and $i,j$ the two nonmerged quartet labels. Put $a=c_{Kij}$, $w=c_{Kuv}$, and $b_i=c_{Kiu}+c_{Kiv}$, with $b_j$ defined similarly. The input quartet $i,j,u,v$ gives
319
+ \[
320
+ (aw)^{1/q}\le(c_{Kiu}c_{Kjv})^{1/q}+(c_{Kiv}c_{Kju})^{1/q}.
321
+ \]
322
+ Using $(x+y)^q\le2^{r-1}(x^q+y^q)$ for $q\ge1$, and subadditivity for $q\le1$, yields
323
+ \[
324
+ 2aw\le2^r(c_{Kiu}c_{Kjv}+c_{Kiv}c_{Kju})\le2^r b_i b_j.
325
+ \]
326
+ The output products are $2aw,b_i b_j,b_i b_j$. Thus they form a null list already at $p=r$.
327
+
328
+ \textbf{Case $(1,0)$: the only parameter-doubling case.}
329
+ Write $\alpha=C+z$, $|C|=d-3$, and let the quartet be $i,j,k,\ell$. Define
330
+ \[
331
+ a^h_{ij}=c_{Cijh}^{1/q}\quad(h=u,v),\qquad w_i=c_{Ciuv}^{1/q}>0.
332
+ \]
333
+ Apply the input weak relation with common index $C\cup\{u\}$ and quartet $i,j,k,v$. One of its triangle inequalities is
334
+ \[
335
+ a^u_{ij}w_k\le a^u_{ik}w_j+a^u_{jk}w_i.
336
+ \]
337
+ The analogous statement holds for $v$. Hence
338
+ \[
339
+ d^h_{ij}=\frac{a^h_{ij}}{w_iw_j},\qquad d^h_{ii}=0,
340
+ \]
341
+ are metrics on the unmerged labels under consideration. Positivity is essential here and is part of the theorem's hypotheses. For $q\ge1$ their pointwise $\ell_q$ aggregate
342
+ \[
343
+ D_{ij}=((d^u_{ij})^q+(d^v_{ij})^q)^{1/q}
344
+ \]
345
+ is a metric by Minkowski. For $q\le1$, each $(d^h)^q$ is a metric, so $D_{ij}=(d^u_{ij})^q+(d^v_{ij})^q$ is a metric. In both cases the merged coefficients satisfy
346
+ \begin{equation}\label{eq:metricfactor}
347
+ b_{ij}:=c_{Ciju}+c_{Cijv}=(w_iw_j)^qD_{ij}^{r}.
348
+ \end{equation}
349
+ At parameter $p=2r$, the root product associated with a pairing of the quartet equals
350
+ \[
351
+ (b_{ij}b_{k\ell})^{1/(2r)}
352
+ =(w_iw_jw_kw_\ell)^{q/(2r)}\sqrt{D_{ij}D_{k\ell}}.
353
+ \]
354
+ The common positive multiplier cancels. Lemma~\ref{lem:snowflake} supplies precisely all three required inequalities.
355
+
356
+ \textbf{Case $(1,1)$.} Write $\alpha=C+z$ and use quartet $i,j,k,z$. Set $W_i=c_{Ciuv}$ and $B_{jk}=c_{Cjku}+c_{Cjkv}$. The three output products are
357
+ \[
358
+ 2W_iB_{jk},\quad2W_jB_{ik},\quad2W_kB_{ij}.
359
+ \]
360
+ For each fixed $h=u,v$, the products $W_i c_{Cjkh}$, $W_j c_{Cikh}$, $W_k c_{Cijh}$ form an input weak relation with common index $C\cup\{h\}$. Aggregate the two lists using Lemma~\ref{lem:minkowski} and discard the shared factor two. This proves nullity at $p=r$.
361
+
362
+ \textbf{Case $(2,0)$.} The common index has the form $C+2z$. Each coefficient is twice an original coefficient with $u,v$ present. All three products are multiplied by four, and the input weak relation with common index $C\cup\{u,v\}$ applies.
363
+
364
+ These six cases exhaust all admissible tests. In degree two, $|\alpha|=0$, so only the first three cases occur and $p=r$ suffices throughout. Monotonicity proves the assertion for larger $p$.
365
+ \end{proof}
366
+
367
+ \begin{theorem}[Arbitrary matroid support]\label{thm:zeros}
368
+ The normalized output of one merge of any weak $\T_q$ matroid coefficient array is weak over $\T_{2\max\{q,1\}}$. In degree two, $\T_{\max\{q,1\}}$ suffices, with strong output as well.
369
+ \end{theorem}
370
+ \begin{proof}
371
+ The output support is M-convex by Lemma~\ref{lem:support} below. The five elementary cases in the preceding proof allow zeros unchanged. Consider its critical case $(1,0)$ and fix its quartet. Retain the notation $a^h_{ij}=c_{Cijh}^{1/q}$ and $w_i=c_{Ciuv}^{1/q}$.
372
+
373
+ If every $w_i>0$, the preceding metric argument applies directly with pseudometrics: its divisions are valid, and vanishing distances do not affect Minkowski or Lemma~\ref{lem:snowflake}. Hence assume that at least one weight vanishes.
374
+
375
+ Suppose some $w_i$ is positive. The weighted triangle inequalities imply that an edge between two zero-weight vertices has $a^h_{ij}=0$: use a positive third vertex. They also imply that for a zero-weight vertex $i$ and positive-weight vertices $j,k$, $a^h_{ij}/w_j=a^h_{ik}/w_k=:A_{i,h}$, by applying both directions of the same inequality. If at least two weights vanish, the output products are either all zero or $(0,B,B)$, so are null at every parameter.
376
+
377
+ If exactly one weight, say $w_i$, vanishes, put $M=A_{i,u}^q+A_{i,v}^q$. For positive vertices $j$ we have $b_{ij}=w_j^q M$. On the other three vertices the aggregate metric $D$ in \eqref{eq:metricfactor} is defined as before, allowing zero distances. If $M=0$ all output products vanish. Otherwise each output root product at parameter $2r$ is a common positive multiple of one of $\sqrt{D_{jk}},\sqrt{D_{j\ell}},\sqrt{D_{k\ell}}$. These form a triangle, since square roots of metric distances satisfy the triangle inequality. Lemma~\ref{lem:snowflake} also holds for pseudometrics by the same proof.
378
+
379
+ It remains to treat $w_i=0$ at all four vertices. The input weak relation with common index $C\cup\{i\}$ and quartet $j,k,u,v$ now has two nonzero candidate products, and forces
380
+ \[
381
+ c_{Ciju}c_{Cikv}=c_{Cijv}c_{Ciku}.
382
+ \]
383
+ Thus nonzero coefficient two-vectors on adjacent edges are proportional. Suppose two disjoint nonzero edge vectors $v_{ij}=(c_{Ciju},c_{Cijv})$ and $v_{k\ell}$ were not proportional. Every bridge vector $v_{ik},v_{i\ell},v_{jk},v_{j\ell}$ is adjacent to both and must be zero. Choose nonzero coordinates $h,g\in\{u,v\}$ to obtain actual matroid bases $C\cup\{i,j,h\}$ and $C\cup\{k,\ell,g\}$. Basis exchange removing $i$ from the first basis is impossible: exchanging with $k$ or $\ell$ gives a zero bridge coefficient, and exchanging with $g\ne h$ gives $c_{Cjuv}=0$. This contradicts the support matroid's basis-exchange axiom.
384
+
385
+ All six nonnegative edge vectors therefore lie on one ray (zero vectors allowed). If every vector is zero there is nothing to prove. Otherwise choose a component $h$ positive on that ray. There is a single scalar $\lambda>0$ such that $b_{ij}=\lambda c_{Cijh}$ for every edge. The output relation is a rescaled input weak relation, valid already at $q$ and hence at $2r$. This completes all zero cases without a density assumption. Degree two never uses this critical case, so its earlier bound holds for every rank-two matroid.
386
+ \end{proof}
387
+
388
+ \begin{remark}[Scope of the repair]
389
+ The output is weak; strong output in degree at least three is not proved. The arbitrary-support extension uses the matroid support axiom in its final zero case; no positivity or strong-input hypothesis is needed. Weighted two-variable merges are covered by absorbing their positive weights into diagonal input rescalings.
390
+ \end{remark}
391
+
392
+ \section{Sharp thresholds and asymptotically optimal parameter cost}\label{sec:optimal}
393
+ \subsection{Determinants as certified strong inputs}
394
+ \begin{lemma}\label{lem:det}
395
+ For a real rank-$d$ matrix $A$ and any $q>0$, $c_S=|\det A_S|^q$ is strong over $\T_q$ on its column matroid.
396
+ \end{lemma}
397
+ \begin{proof}
398
+ The signed determinant Pl\"ucker relation sums to zero. The absolute value of each term is at most the sum of the others by the ordinary triangle inequality. These absolute values are exactly the $q$th roots of the products in \eqref{eq:strong}.
399
+ \end{proof}
400
+
401
+ Take four planar points $u_i\in\R^2$, their affine lifts $(u_i,1)$, and two additional columns $e_1,e_2\in\R^3$. After merging the last two labels,
402
+ \begin{equation}\label{eq:planar}
403
+ b_{ij}=|u_{i,1}-u_{j,1}|^q+|u_{i,2}-u_{j,2}|^q.
404
+ \end{equation}
405
+ This directly verifies the required output coefficients; no metric theorem is needed for the lower bounds below.
406
+
407
+ \begin{proposition}[Necessary output parameter]\label{prop:lower}
408
+ Any universal parameter $p$ for one merge in degree at least three, even restricted to positive uniform strong $\T_q$ inputs, must satisfy
409
+ \begin{equation}\label{eq:lower}
410
+ p\ge\max\{2,2(q-1)\}.
411
+ \end{equation}
412
+ Moreover, in each fixed degree $d\ge3$ and size $n\ge d+3$, the infimal universal parameter is strictly greater than $q$.
413
+ \end{proposition}
414
+ \begin{proof}
415
+ For the planar unit square, \eqref{eq:planar} gives adjacent coefficients one and diagonal coefficients two, independently of $q$. The three output products are $(1,4,1)$. The output $\T_p$ relation holds exactly when $4^{1/p}\le2$, or $p\ge2$.
416
+
417
+ For the coordinate diamond $(1,0),(0,1),(-1,0),(0,-1)$ the three products are $(4,4^q,4)$. When $q>1$, the necessary inequality $4^{q/p}\le2\cdot4^{1/p}$ is equivalent to $p\ge2(q-1)$; for $q\le1$ this bound is vacuous.
418
+
419
+ These matrices may have zero minors. Whenever a proposed bound is violated, the failed output inequality is strict. A sufficiently small generic matrix perturbation makes every maximal minor nonzero and preserves this strict failure, while Lemma~\ref{lem:det} certifies strong input at the fixed $q$. To realize every $d,n$, first add $d-3$ identity columns and rows, then any extra columns, and perturb generically. The selected quartet minors are nonzero before perturbation, so the same failure persists.
420
+
421
+ Finally, Theorem~\ref{thm:explicit} provides, at each fixed $q,d,n$, a witness with a positive output defect at parameter $q$. That witness continues to fail for all output parameters in some interval $[q,q+\eps]$ by continuity. Thus the infimum of universal working parameters is at least $q+\eps>q$.
422
+ \end{proof}
423
+
424
+ \subsection{Degree two is completely determined}
425
+ \begin{theorem}[Exact degree-two answer]\label{thm:d2}
426
+ For degree-two matroid input, the infimal universal output parameter for one merge is
427
+ \[
428
+ P_2(q)=\max\{q,1\}.
429
+ \]
430
+ The output is both weak and strong at this parameter. Consequently the same-parameter threshold in degree two is exactly $q_2=1$.
431
+ \end{theorem}
432
+ \begin{proof}
433
+ The upper bound is Theorems~\ref{thm:repair} and \ref{thm:zeros}. In rank two every strong relation has at most three summands, so weak and strong coincide. Equivalently, polarize the output by replacing the capacity-two coordinate $z$ with two clones. Its coefficient on the two clones is $\rho_{2z}$ and its coefficients on a clone and $i$ are $\rho_{zi}$. Every rank-two strong test is a three-term quartet, including the repeated-$z$ test already checked.
434
+
435
+ For $p\ge q$ necessity use four real columns $(1,0),(0,1),(1,1),(1,-1)$: their Pl\"ucker product magnitudes are $(2,1,1)$. After raising minors to $q$, the products are $(2^q,1,1)$, which require $p\ge q$. Add two very small generic columns and merge those columns; the untouched quartet is unchanged and all input minors can be nonzero.
436
+
437
+ For $p\ge1$ necessity take the five real columns
438
+ \[
439
+ e_1,\quad e_2,\quad e_1+e_2,\quad e_1,\quad e_2
440
+ \]
441
+ and merge the last two. On the unmerged labels $1,2,3$ the coefficients are all one. The merged coefficients are $b_1=b_2=1,b_3=2$, for every input exponent $q>0$. The output quartet products are $(2,1,1)$ and require $p\ge1$. A small generic perturbation gives positive uniform input while retaining strict failure for any fixed proposed $p<1$. Extra columns can be added if desired. The two lower bounds and the upper bound agree.
442
+ \end{proof}
443
+ Degree one has only linear coefficients; identification preserves it at every parameter.
444
+
445
+ \subsection{Exact low-parameter repair and the factor two}
446
+ For fixed $d\ge3,n\ge d+3$, let $P^{\rm str}_{d,n}(q)$ be the infimum of parameters $p$ for which every positive uniform strong $\T_q$ input has normalized weak $\T_p$ output after one specified merge. Define $P^{\rm w}_{d,n}(q)$ similarly for weak input. Monotonicity makes the universal working sets upper intervals, and Theorem~\ref{thm:repair} makes them nonempty.
447
+
448
+ \begin{corollary}[Two-sided aggregation law]\label{cor:cost}
449
+ For $d\ge3,n\ge d+3$,
450
+ \[
451
+ P^{\rm str}_{d,n}(q)\le P^{\rm w}_{d,n}(q)\le2\max\{q,1\},
452
+ \]
453
+ \[
454
+ P^{\rm str}_{d,n}(q)\ge\max\{2,2(q-1)\},
455
+ \qquad P^{\rm str}_{d,n}(q)>q.
456
+ \]
457
+ In particular,
458
+ \[
459
+ P^{\rm str}_{d,n}(q)=P^{\rm w}_{d,n}(q)=2\quad(0<q\le1),
460
+ \]
461
+ and
462
+ \[
463
+ \lim_{q\to\infty}\frac{P^{\rm str}_{d,n}(q)}q
464
+ =\lim_{q\to\infty}\frac{P^{\rm w}_{d,n}(q)}q=2.
465
+ \]
466
+ For $q\ge1$, the additive gap between the stated upper and lower bounds is at most two.
467
+ \end{corollary}
468
+ \begin{proof}
469
+ Combine Theorem~\ref{thm:repair} and Proposition~\ref{prop:lower}. For $q\ge2$, the bounds $2q-2$ and $2q$ squeeze the ratios to two and have difference two. For $1\le q\le2$, the lower bound two and upper bound $2q$ also have difference at most two.
470
+ \end{proof}
471
+ The full higher-degree optimum remains undetermined by the elementary arguments here. Section~\ref{sec:peaks} strengthens the lower bounds and gives a quantitative obstruction at $q=2$.
472
+
473
+ \section{The exact minimal obstruction boundary}\label{sec:minimal}
474
+ The universal failure in rank at least three first appears at corank three. Below it, the full normalized output retains strong compatibility in a precisely determined parameter range.
475
+
476
+ For clarity, strong compatibility of the one-merge output can be tested by its natural polarization: replace the capacity-two variable $z$ by two labeled clones, and assign to a $d$-subset the normalized coefficient $\rho_\beta$ with the same count vector $\beta$. The clone array has $n$ ground labels. Weak clone relations are exactly the bounded-multiplicity quartet tests of the output, with the usual zero or equal-pair tautologies. Its support is a matroid. For a direct proof, take two clone bases with count vectors $x,y$ and a clone in the first but not the second, in block $h$. If $x_h\le y_h$, swap it with an unused clone of the second basis in the same block. If $x_h>y_h$, M-convex exchange supplies a deficit block $j$ and valid exchanged counts; choose a clone in the second but not the first in that block. This verifies ordinary basis exchange for the lift. Lemma~\ref{lem:support} proves the needed M-convexity.
477
+
478
+ \begin{theorem}[Complete one-merge preservation classification]\label{thm:minimal}
479
+ Fix $q>0$, rank $d\ge1$, and $n\ge\max(d,2)$, with a specified pair to be merged. Universal same-parameter preservation from strong input to strong full normalized output holds exactly in the regimes of Table~\ref{tab:threshold}. Replacing ``strong output'' by ``weak output'' gives the identical classification. Every excluded regime admits a uniform positive input counterexample.
480
+ \end{theorem}
481
+ \begin{proof}
482
+ Degree one is immediate. For $d\ge2$ and $n\le d+1$, the polarized output has corank at most one. A strong relation has
483
+ \[
484
+ |J\setminus I|\le n-d+1\le2
485
+ \]
486
+ terms, hence is zero or a tautological equal pair. Thus every parameter works.
487
+
488
+ Suppose $n=d+2$. The critical metric case $(\alpha_z,k_z)=(1,0)$ in Theorem~\ref{thm:repair} requires a common set of $d-3$ labels, four quartet labels, and the two merged labels: $d+3$ distinct source elements. It is impossible here. All five remaining cases work already at $p=\max\{q,1\}$, including their zero cases. For $q\ge1$ the parameter stays $q$. The polarized output now has corank two, so all its strong relations have at most three terms and weak equals strong.
489
+
490
+ For $d=2$ and any $n\ge4$, Theorem~\ref{thm:congruence} gives same-parameter strong preservation whenever $q\ge1$.
491
+
492
+ For the converse at $q<1$, choose $M>0$ and the rank-two real columns
493
+ \[
494
+ i=(1,0),\quad j=(0,1),\quad
495
+ u=(1,M^{1/q}),\quad v=(M^{1/q},-1).
496
+ \]
497
+ Their absolute-minor $q$-power coefficients are strong by Lemma~\ref{lem:det}. Merging $u,v$ gives the normalized output product list
498
+ \[
499
+ 2(1+M^{2/q})^q,\qquad (M+1)^2,\qquad(M+1)^2.
500
+ \]
501
+ The ratio of the first product to the other two tends to $2$ as $M\to\infty$. Nullity at parameter $q$ would require that ratio to be at most $2^q<2$, so sufficiently large $M$ gives strict failure. All six source minors are nonzero. Adding $d-2$ identity coloops gives rank $d$ on $d+2$ elements; a sufficiently small generic real matrix perturbation makes every maximal minor nonzero while preserving the strict failed relation. For degree two, additional generic columns cover every $n>4$ as well.
502
+
503
+ Finally, if $d\ge3,n\ge d+3$, Theorem~\ref{thm:explicit} supplies a strict positive uniform counterexample at every $q>0$. This exhausts the table and proves both classifications.
504
+ \end{proof}
505
+
506
+ \begin{example}[Smallest exact failure below one]
507
+ At $q=1/2$ and $M=8$, the four real columns above are
508
+ $(1,0),(0,1),(1,64),(64,-1)$. The nontrivial input root products are $(4097,4096,1)$, so the input is strong. The merged output root products at $q=1/2$ are
509
+ \[
510
+ 16388,\qquad6561,\qquad6561.
511
+ \]
512
+ The excess is the integer $3266>0$. Thus failure already occurs on $U_{2,4}$, the smallest possible nontrivial ground set in degree two.
513
+ \end{example}
514
+
515
+ \section{Two peaks: exact weak versus strong compatibility}\label{sec:peaks}
516
+ Fix $d\ge3,n\ge d+3$ and $|C|=d-3$ disjoint from $[6]$. Set $U=C\cup135$ and $V=C\cup246$. These two exceptional bases meet in $d-3$ elements.
517
+
518
+ \begin{theorem}[Exact two-peak classification]\label{thm:twopeak}
519
+ Let $a,b\ge1$, define root coefficients $a_U=a$, $a_V=b$, and $a_S=1$ otherwise, and put $c_S=a_S^q$. Then
520
+ \[
521
+ c\text{ is weak over }\T_q\quad\Longleftrightarrow\quad a\le2,\ b\le2,
522
+ \]
523
+ \[
524
+ c\text{ is strong over }\T_q\quad\Longleftrightarrow\quad
525
+ a\le2,\ b\le2,\ ab\le3.
526
+ \]
527
+ All genuine weak inequalities are strict when $a,b<2$, and all genuine strong inequalities are strict when additionally $ab<3$.
528
+ \end{theorem}
529
+ \begin{proof}
530
+ A weak relation has union size $d+2$, whereas $U\cup V$ has size $d+3$. Thus it contains at most one peak, and its root products are $(1,1,1)$, $(a,1,1)$, or $(b,1,1)$. Both peak types occur; this proves the weak equivalence and its strictness statement.
531
+
532
+ For a four-term strong relation, exceptional factors cannot lie in different terms: factors in different terms intersect in either $d-1$ or $d-2$ elements, whereas $|U\cap V|=d-3$. If both peaks occur they pair together and give $(ab,1,1,1)$. Such a relation exists, with $I=C\cup\{1,3\}$ and $J=C\cup\{2,4,5,6\}$. It forces $ab\le3$. Relations with a single peak are harmless under $a,b\le2$. A relation with at least five terms has at most two exceptional terms, retaining at least three unit terms; its largest term is at most two. Both peaks cannot occupy one product in this case, since the intersection of the two factors of a product would have to have size $d+1-m<d-3$. Therefore all these relations are strict. This proves sufficiency and necessity.
533
+ \end{proof}
534
+
535
+ After merging $5,6$, four relevant coefficients equal two and the other two equal $1+a^q$ and $1+b^q$. Their product list is
536
+ \begin{equation}\label{eq:peakoutput}
537
+ 4,\qquad(1+a^q)(1+b^q),\qquad4.
538
+ \end{equation}
539
+
540
+ \begin{corollary}[Sharper weak repair lower bound]\label{cor:weaklower}
541
+ For every $d\ge3,n\ge d+3$,
542
+ \[
543
+ P^{\rm w}_{d,n}(q)\ge
544
+ \max\left\{2,\;2\log_2\frac{1+2^q}{2}\right\}.
545
+ \]
546
+ For strong inputs one also has
547
+ \[
548
+ P^{\rm str}_{d,n}(q)\ge
549
+ \log_2\frac{(1+2^q)(1+(3/2)^q)}4.
550
+ \]
551
+ \end{corollary}
552
+ \begin{proof}
553
+ Take $a=b=2$ for the weak assertion, and $a=2,b=3/2$ for the strong assertion, and solve the null inequality for \eqref{eq:peakoutput}. The universal lower bound two is Proposition~\ref{prop:lower}. Lowering the peaks slightly makes the required input inequalities strict while preserving every strict output failure, so the bounds also hold when testing only strict inputs.
554
+ \end{proof}
555
+ The weak bound exceeds $2(q-1)$ strictly for every $q>0$. It forces $P^{\rm w}_{d,n}(q)>2$ whenever $q>\log_2 3$. No matching upper curve is asserted here.
556
+
557
+ \begin{example}[Three integer levels at parameter two]
558
+ On every $U_{d,n}$ in this section, set $c_U=23$, $c_V=14$, and every other coefficient to $6$. After removing the common root scale $\sqrt6$, the peaks are $\sqrt{23/6}<2$ and $\sqrt{14/6}<2$, with product $\sqrt{322}/6<3$. The input is strict strong by Theorem~\ref{thm:twopeak}. Its merged product list is $(580,144,144)$, which fails $\T_2$ because
559
+ \[
560
+ 580-(12+12)^2=4>0.
561
+ \]
562
+ \end{example}
563
+
564
+ \begin{theorem}[A quantitative strong-input barrier at two]\label{thm:twentyninths}
565
+ For every $d\ge3,n\ge d+3$,
566
+ \[
567
+ P^{\rm str}_{d,n}(2)>\frac{20}{9}.
568
+ \]
569
+ \end{theorem}
570
+ \begin{proof}
571
+ Use the three exceptional bases $U=C\cup135$, $Z_1=C\cup126$, $Z_2=C\cup346$. Assign
572
+ \[
573
+ c_U=3999,\quad c_{Z_1}=c_{Z_2}=1,\quad
574
+ c_S=5000\ (6\in S\text{ otherwise}),\quad
575
+ c_S=1000\ (6\notin S\text{ otherwise}).
576
+ \]
577
+ This is Lemma~\ref{lem:uniform} at $q=2$, root peak $\sqrt{3999/1000}<2$, root low $1/\sqrt{5000}<1/2$, followed by scaling label six by five and globally by $1000$. Hence all genuine strong input inequalities are strict.
578
+
579
+ The merged coefficients are $g_{12}=g_{34}=1001$, $g_{14}=g_{23}=g_{24}=6000$, and $g_{13}=8999$. Divide their three products by $10^6$ to obtain
580
+ \[
581
+ A=\frac{26997}{500},\qquad B=36,\qquad D=\frac{1002001}{1000000}.
582
+ \]
583
+ Exact integer-power comparisons give
584
+ \[
585
+ A^9>\left(\frac{6019}{1000}\right)^{20},\quad
586
+ 36^9<\left(\frac{5016}{1000}\right)^{20},\quad
587
+ D^9<\left(\frac{1001}{1000}\right)^{20}.
588
+ \]
589
+ For a short manual or executable check, multiply each inequality by its positive denominators. For example, the first is exactly
590
+ $26997^9\,1000^{20}>6019^{20}\,500^9$; the other two are equally direct integer comparisons. They imply
591
+ \[
592
+ A^{9/20}-B^{9/20}-D^{9/20}>\frac{2}{1000}>0.
593
+ \]
594
+ Thus the fixed strict input fails at output parameter $20/9$. Continuity in the output parameter gives failure on a nonempty interval above $20/9$, proving the strict inequality for the infimal universal repair parameter. All selected relations are unchanged by extra ground labels.
595
+ \end{proof}
596
+
597
+ \section{Exact classification of weighted squarefree aggregation}\label{sec:classification}
598
+ Fix $d\ge3$ and $q>0$. Let an operator substitute
599
+ \[
600
+ x_i\longmapsto w_i y_{\phi(i)},\qquad w_i\ge0,
601
+ \]
602
+ and then take the squarefree part. Only the active labels $E_+=\{i:w_i>0\}$ matter. Let $m=|\phi(E_+)|$. Write $A_{\phi,w}$ for this squarefree operator. A zero output is allowed; otherwise its coefficient array has rank $d$.
603
+
604
+ \begin{theorem}[Weighted set-map classification]\label{thm:classification}
605
+ The following are equivalent:
606
+ \begin{enumerate}
607
+ \item For every strong $\T_q$ rank-$d$ input, $A_{\phi,w}$ is zero or strong over $\T_q$.
608
+ \item For every strong $\T_q$ rank-$d$ input, $A_{\phi,w}$ is zero or weak over $\T_q$.
609
+ \item Either $\phi|_{E_+}$ is injective, or $m\le d+1$.
610
+ \end{enumerate}
611
+ When condition 3 fails, failure can be witnessed by a positive uniform input on the original ground set, with every genuine input inequality strict.
612
+ \end{theorem}
613
+ \begin{proof}
614
+ Condition 1 implies 2. If the active map is injective, the operation is deletion, positive rescaling, and relabeling; strong representability is preserved, unless every rank-$d$ term is deleted. If $m\le d+1$, any nonzero rank-$d$ array on these labels is strong: for $m=d$ it has at most one basis, and for $m=d+1$ every strong relation has only its two identical terms. Its support is a matroid, since complements of the possible bases are the nonloops of a rank-one matroid. For $m<d$ the squarefree output is zero. This proves 3 implies 1.
615
+
616
+ Suppose instead that $\phi$ has a collision and $m\ge d+2$. Choose two active labels in one fiber and call them $5,6$. Choose four other distinct fibers and representatives $1,2,3,4$. Choose a further $d-3$ distinct fibers with representatives $C$. The total number of used output labels is $d+2$, so these choices are possible.
617
+
618
+ Apply Theorem~\ref{thm:explicit} on the entire original ground set with this distinguished configuration. Rescale each active variable by $1/w_i$. Additionally rescale every active variable outside the chosen $d+3$ labels by a parameter $\lambda>0$. Input positivity and strictness are preserved. In the limit $\lambda\to0$, the six output coefficients of the selected weak relation converge to \eqref{eq:six}; contributions involving any additional element of the same output fibers vanish. Inactive variables contribute nothing. The limiting output relation fails by at least $L/2$, so it still fails for all sufficiently small positive $\lambda$. This contradicts condition 2 and proves necessity. Every input coefficient remains positive for such $\lambda$, as asserted.
619
+ \end{proof}
620
+
621
+ \begin{remark}[Why the squarefree qualifier is necessary]
622
+ The low-output-count exception concerns rank-$d$ arrays on at most $d+1$ labels. The full polynomial may have repeated monomials and additional weak tests. Theorem~\ref{thm:classification} is not a classification of all full-polynomial linear preservers, nor of arbitrary matrices of linear forms.
623
+ \end{remark}
624
+
625
+ \begin{corollary}[A guaranteed rule for repeated squarefree merging]
626
+ Start with positive weak data on $U_{d,n}$. Perform $k$ binary identifications, taking the squarefree shadow after each, and assume the number of labels remains at least $d$. The resulting array is weak over $\T_p$ for
627
+ \[
628
+ p=2^k\max\{q,1\}.
629
+ \]
630
+ For $d=2$ the better parameter $\max\{q,1\}$ suffices after any number of such steps.
631
+ \end{corollary}
632
+ \begin{proof}
633
+ Each squarefree shadow remains positive on the new uniform support, and its weak tests are among the full output weak tests just proved. Iterate Theorem~\ref{thm:repair}; in degree two iterate $r\mapsto\max\{r,1\}$ instead.
634
+ \end{proof}
635
+ The exponential bound is a safe composition bound, not an optimal cost for a partition or for multiple merges.
636
+
637
+ \section{A robust Lorentzian obstruction to strong compatibility}\label{sec:lorentzian}
638
+ The preceding repair results distinguish weak and strong output. This section gives an independent explicit obstruction: even a Lorentzian cubic can fail strong compatibility in the same parameter that controls its local quadratic geometry.
639
+
640
+ For a positive multi-affine cubic on six labels, we use the standard Lorentzian criterion: each quadratic derivative has a Hessian with at most one positive eigenvalue. Uniform support supplies the required M-convexity. This is the usual derivative characterization in \cite{BH}. The active Hessian omits the zero row and column of the differentiated variable. Signature $(1,4)$ of all six active Hessians implies membership in the relative interior of the Lorentzian $U_{3,6}$ stratum.
641
+
642
+ \begin{theorem}[Interior Lorentzian obstruction]\label{thm:lorentz}
643
+ Partition $[6]$ as $P=\{1,3\}$, $Q=\{2,4,6\}$, and $\{s\}=\{5\}$. For $t>1$ and $0<\delta<1$, put
644
+ \[
645
+ a=3-\delta,\qquad h=4-t^{-2}-\delta.
646
+ \]
647
+ Assign the twenty cubic coefficients by the following exhaustive composition classes:
648
+ \begin{center}
649
+ \begin{tabular}{lrrrrrr}
650
+ \toprule
651
+ Class & $PPs$ & $QQQ$ & $PPQ$ & $sQQ$ & $PsQ$ & $PQQ$\\
652
+ Number & 1 & 1 & 3 & 3 & 6 & 6\\
653
+ Coefficient & $a$ & $h$ & 1 & 1 & $t$ & $t$\\\bottomrule
654
+ \end{tabular}
655
+ \end{center}
656
+ Every active derivative Hessian has signature $(1,4)$. The strong relation indexed by $I=P$ and $J=Q\cup\{s\}$ has product list $(ah,1,1,1)$. Therefore any parameter $p$ containing all Lorentzian $U_{3,6}$ coefficient arrays among strong $\T_p$ arrays must satisfy
657
+ \[
658
+ p\ge\log_3 12.
659
+ \]
660
+ \end{theorem}
661
+ \begin{proof}
662
+ All coefficients are positive. Symmetry reduces the six active Hessians to three types; the quotient matrices below use orthonormal constant vectors on their symmetry blocks.
663
+
664
+ For differentiation in $s$, the quotient on $P,Q$ is
665
+ \[
666
+ \begin{pmatrix}a&\sqrt6t\\\sqrt6t&2\end{pmatrix}.
667
+ \]
668
+ Its determinant $2a-6t^2$ is negative, giving one positive and one negative eigenvalue. The other three modes have eigenvalues $-a,-1,-1$.
669
+
670
+ For differentiation in a member of $P$, the symmetry blocks have sizes $1,1,3$, and the quotient is
671
+ \[
672
+ \begin{pmatrix}
673
+ 0&a&\sqrt3\\ a&0&\sqrt3t\\ \sqrt3&\sqrt3t&2t
674
+ \end{pmatrix}.
675
+ \]
676
+ Its determinant is $2at\delta>0$ and the sum of its principal two-by-two minors is $-a^2-3-3t^2<0$. Thus its three nonzero eigenvalues have one positive and two negative signs: their product is positive, while the all-positive possibility is excluded by the negative sum of pairwise products. The other two modes are $-t,-t$.
677
+
678
+ For differentiation in a member of $Q$, the symmetry blocks have sizes $2,2,1$, and the quotient is
679
+ \[
680
+ \begin{pmatrix}
681
+ 1&2t&\sqrt2t\\ 2t&h&\sqrt2\\ \sqrt2t&\sqrt2&0
682
+ \end{pmatrix}.
683
+ \]
684
+ Its determinant is $2t^2\delta>0$ and the sum of its principal two-by-two minors is $h-6t^2-2<0$. Again it has one positive and two negative eigenvalues. The other two modes are $-1,-h$. This proves signature $(1,4)$ in every case.
685
+
686
+ In the specified strong relation, the $s$ term is $c_{PPs}c_{QQQ}=ah$; each of the other three terms is $c_{PPQ}c_{sQQ}=1$. Strong nullity requires $(ah)^{1/p}\le3$, so $p\ge\log_3(ah)$. Let $\delta\downarrow0$ and $t\to\infty$ to obtain $p\ge\log_3 12$.
687
+ \end{proof}
688
+
689
+ \begin{corollary}[A robust four-level integer certificate]\label{cor:lorentzbox}
690
+ Take $t=2$, $\delta=1/4$, and multiply every coefficient by four. The six coefficient classes become
691
+ \[
692
+ 11,\quad14,\quad4,\quad4,\quad8,\quad8.
693
+ \]
694
+ This Lorentzian array fails strong $\T_2$; moreover every coefficient perturbation of absolute size at most $1/1000$ retains both properties within the multi-affine coefficient space.
695
+ \end{corollary}
696
+ \begin{proof}
697
+ The strong product list is $(154,16,16,16)$, and $154>144=(4+4+4)^2$ proves failure. Exact active-Hessian determinants, in variable order, are
698
+ \[
699
+ 11264,\ 7168,\ 11264,\ 7168,\ 52096,\ 7168.
700
+ \]
701
+ Each maximum absolute row sum is at most $35$. All eigenvalues therefore have absolute value at most $35$, while the determinant product implies every absolute eigenvalue is at least $7168/35^4$. A perturbation of the twenty coefficients by at most $\eps=1/1000$ changes each active Hessian by operator norm at most $4\eps$, because each row has only four off-diagonal entries. The exact inequality
702
+ $4/1000<7168/35^4$ prevents any eigenvalue from crossing zero, so the active signatures remain $(1,4)$.
703
+
704
+ The worst permitted strong-product comparison still has positive squared excess:
705
+ \[
706
+ (11-\eps)(14-\eps)-9(4+\eps)^2
707
+ =10-97\eps-8\eps^2>0.
708
+ \]
709
+ All coefficients remain positive. This gives the stated explicit relative open box.
710
+ \end{proof}
711
+ This is a lower-bound theorem for a distinct inclusion problem. It does not establish finiteness or a matching upper bound for the Lorentzian-to-strong parameter.
712
+
713
+ \section{Concrete certificates and verification}\label{sec:verification}
714
+ The package contains the complete source, an executable verifier, and its JSON results. Universal real-parameter statements are established by the proofs above. Finite computations are certificates for selected constructions and checks against transcription mistakes.
715
+
716
+ \subsection{A strict rational witness at $q=1$}
717
+ On $U_{3,6}$ take root coefficients $a_{135}=199/100$, $a_{126}=a_{346}=1/4$, and all others one; rescale label six by $s=1/4$. Lemma~\ref{lem:uniform} proves strict strong input. The relevant output coefficients are
718
+ \[
719
+ g_{12}=g_{34}=17/16,\quad
720
+ g_{14}=g_{23}=g_{24}=5/4,\quad
721
+ g_{13}=56/25.
722
+ \]
723
+ The failed triangle defect, with no numerical roots required, is
724
+ \[
725
+ \frac{56}{25}\frac54-\left(\frac54\right)^2-
726
+ \left(\frac{17}{16}\right)^2=\frac{139}{1280}>0.
727
+ \]
728
+ This example is convenient for an independent manual check of strictness, merging, and exact rational failure.
729
+
730
+ \subsection{A strict integer witness at $q=2$}
731
+ An explicit integer construction is
732
+ \[
733
+ c_{135}=4095,\quad c_{126}=c_{346}=8,\quad
734
+ c_S=64\ (6\in S\text{ otherwise}),\quad
735
+ c_S=1024\ (6\notin S\text{ otherwise}).
736
+ \]
737
+ Its six merged coefficients are $g_{12}=g_{34}=1032$, $g_{14}=g_{23}=g_{24}=1088$, and $g_{13}=4159$. Failure is certified by
738
+ \[
739
+ 4159\cdot1088-(1088+1032)^2=30592>0.
740
+ \]
741
+ The input is obtained from Lemma~\ref{lem:uniform} with root peak $\sqrt{4-1/1024}$ and low $1/\sqrt8$, followed by label-six scaling $1/16$ and global scaling $1024$. Exact enumeration provides a second certificate.
742
+
743
+ \subsection{What is checked}
744
+ The delivered scripts separate exact certificates from numerical checks.
745
+ \begin{itemize}
746
+ \item The core verifier enumerates all strong relations in eight rank/size configurations, certifies the rational and integer examples, and checks all six normalized output cases as formal polynomial identities. It also certifies 567 boundary-case relations exactly and checks 264 numerical instances containing 11,424 normalized weak relations.
747
+ \item The congruence verifier checks three formal tensor identities, 16 exact rational matrix instances, 128 numerical substitution/composition instances at 90 digits, the minimal corank example, the two-peak classification in selected ranks and sizes, and the integer levels $6,14,23$.
748
+ \item The strong-parameter verifier certifies every one of 225 indexed input relations for each of two integer witnesses, then checks the output twentieth-power comparisons entirely with integers and fractions.
749
+ \item The Lorentzian verifier proves twelve derivative quotient identities symbolically and checks 372 principal minors exactly at the rational examples. It also verifies the strict integer example and every stated robustness constant.
750
+ \end{itemize}
751
+ All delivered checks pass. Universal real-parameter statements are established by the written proofs, rather than inferred from samples. The scripts are reproducibility certificates, not a proof-assistant formalization.
752
+
753
+ \section{Interpretation, limits, and the next mathematical target}
754
+ The decisive structural distinction is between homogeneous rescaling and additive aggregation. Rescaling acts by one common multiplier in each relation. Aggregation sums coefficients, and a common-index slice becomes a pointwise aggregate of two metrics after a shared normalization. Passing to a square root restores the Ptolemy property of that aggregate; this is the source of the factor two in the repair.
755
+
756
+ The exact quadratic theorem allows arbitrary nonnegative linear maps and their compositions. The cardinality table identifies the complete same-parameter boundary for a full merge. Above that boundary, explicit uniform obstructions and a constructive weak repair coexist. The two-peak family resolves weak versus strong feasibility in a complete parameter region, while the Lorentzian family exhibits a different robust obstruction to strong compatibility.
757
+
758
+ For automated algebra, the practical consequence is a mathematically sound composition rule: a system may lower its claimed output certificate from strong $\T_q$ to weak $\T_{2\max(q,1)}$ after one merge of matroid data. If the same parameter or a strong output is required, new relation checks remain necessary. The paper supplies constant-size regression witnesses and a general operator rejection criterion. It does not measure an AI speedup or establish recursive self-improvement of a model.
759
+
760
+ A focused remaining question is the exact function $P^{\rm str}_{d,n}(q)$ for $q>1$. The current bounds differ by at most two and have the exact asymptotic factor. A second open extension is a comparable repair theorem when the initial input already has repeated coordinates, rather than being multi-affine. A third is higher-degree classification for arbitrary nonnegative linear substitutions rather than weighted set maps; the quadratic case is completely covered by Theorem~\ref{thm:congruence}. None is silently included in the proved assertions.
761
+
762
+ \appendix
763
+ \section{Why merging matroid support is M-convex}
764
+ \begin{lemma}\label{lem:support}
765
+ The image of a matroid's basis indicators under one coordinate merge is M-convex.
766
+ \end{lemma}
767
+ \begin{proof}
768
+ Let $\phi:E\to H$ merge $u,v$ into $z$, leaving other labels distinct. Define the integral monotone submodular function $f(A)=r_M(\phi^{-1}(A))$. Its basis images are exactly the integer vectors
769
+ \[
770
+ x\ge0,\qquad x(H)=d,\qquad x(A)\le f(A)\quad(A\subseteq H).
771
+ \]
772
+ Necessity is immediate. Conversely, every unmerged $x_i\le1$. Let $S$ be the unmerged labels with $x_i=1$, and put $k=x_z\le2$. The inequalities imply that $S$ is independent and $r_M(S\cup\{u,v\})\ge d$. Thus $S$ extends to a basis using exactly $k=d-|S|$ elements of $\{u,v\}$, as required.
773
+
774
+ We prove symmetric exchange for these vectors. Let $x_i>y_i$. Tight sets for $x$ are closed under unions and intersections by submodularity, as are tight sets for $y$. Let $U$ be the union of all $x$-tight sets avoiding $i$ (including the empty set), and $V$ the intersection of all $y$-tight sets containing $i$ (including $H$). Then $i\in V\setminus U$. An exchange $x-e_i+e_j$ preserves all upper inequalities if $j\notin U$; $y+e_i-e_j$ does so if $j\in V$. Integer slack makes every inequality that is not tight harmless.
775
+
776
+ Tightness and submodularity give
777
+ \[
778
+ x(V\setminus U)\le f(U\cup V)-f(U)
779
+ \le f(V)-f(U\cap V)\le y(V\setminus U).
780
+ \]
781
+ Since $x_i>y_i$, some $j\in V\setminus U$ has $x_j<y_j$. Both exchanged vectors are therefore feasible nonnegative integer vectors. This is M-convexity.
782
+ \end{proof}
783
+
784
+ \clearpage
785
+ \begingroup\small
786
+ \begin{thebibliography}{9}
787
+ \bibitem{BHKL} M. Baker, J. Huh, M. Kummer, O. Lorscheid.
788
+ \emph{Lorentzian polynomials and matroids over triangular hyperfields 2: Analytic aspects}.
789
+ arXiv:2607.15375v2, 18 August 2026. In particular Definition 2.21, Proposition 8.5, and Question 8.6.
790
+ \url{https://web.math.princeton.edu/~huh/Triangular2.pdf}.
791
+ \bibitem{BH} P. Br\"and\'en, J. Huh. \emph{Lorentzian polynomials}. Annals of Mathematics 192 (2020), 821--891. \url{https://web.math.princeton.edu/~huh/Lorentzian.pdf}.
792
+ \bibitem{FS} T. Foertsch, V. Schroeder.
793
+ \emph{Hyperbolicity, CAT(-1)-spaces and the Ptolemy inequality}.
794
+ arXiv:math/0605418v2 (2006), Proposition 1.8 and Section 3.
795
+ \url{https://arxiv.org/abs/math/0605418v2}.
796
+ \end{thebibliography}
797
+ \endgroup
798
+ \end{document}
results/VERIFICATION_SUMMARY.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "title": "Ilyoxune: Sharp Laws of Hyperfield Aggregation",
3
+ "status": "all delivered verification scripts passed",
4
+ "scripts": [
5
+ "verify_ilyoxune.py",
6
+ "verify_strengthened_aggregation.py",
7
+ "verify_strong_parameter_bound.py",
8
+ "verify_lorentzian_obstruction.py"
9
+ ],
10
+ "exact": {
11
+ "core_allrank_instances": 8,
12
+ "core_boundary_relations": 567,
13
+ "core_formal_normalization_cases": 6,
14
+ "congruence_instances": 16,
15
+ "formal_tensor_factorizations": 3,
16
+ "weak_two_peak_instances": 4,
17
+ "strong_two_peak_instances": 5,
18
+ "corank_relation_count_instances": 20,
19
+ "strong_parameter_witnesses": 2,
20
+ "strong_parameter_indexed_input_relations_per_witness": 225,
21
+ "lorentzian_symbolic_derivative_certificates": 12,
22
+ "lorentzian_rational_principal_minors": 372
23
+ },
24
+ "numerical": {
25
+ "core_instances": 264,
26
+ "core_output_weak_relations": 11424,
27
+ "congruence_and_composition_instances": 128
28
+ },
29
+ "scope": "Exact and numerical certificates are labeled separately. The manuscript supplies the universal mathematical proofs; finite verification does not establish universal quantifiers."
30
+ }
results/lorentzian_obstruction_results.json ADDED
@@ -0,0 +1,930 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "title": "Exact Lorentzian obstruction to strong T_2",
3
+ "parameter_domain": "real t>1",
4
+ "support": "U_(3,6), all twenty coefficients positive",
5
+ "coefficient_classes": {
6
+ "P": [
7
+ 1,
8
+ 3
9
+ ],
10
+ "Q": [
11
+ 2,
12
+ 4,
13
+ 6
14
+ ],
15
+ "singleton": 5,
16
+ "PPs": "3",
17
+ "QQQ": "4-t^(-2)",
18
+ "PPQ": "1",
19
+ "sQQ": "1",
20
+ "PsQ": "t",
21
+ "PQQ": "t"
22
+ },
23
+ "derivative_certificates": [
24
+ {
25
+ "derivative_label": 1,
26
+ "remaining_labels": [
27
+ 2,
28
+ 3,
29
+ 4,
30
+ 5,
31
+ 6
32
+ ],
33
+ "hessian": [
34
+ [
35
+ "0",
36
+ "1",
37
+ "t",
38
+ "t",
39
+ "t"
40
+ ],
41
+ [
42
+ "1",
43
+ "0",
44
+ "1",
45
+ "3",
46
+ "1"
47
+ ],
48
+ [
49
+ "t",
50
+ "1",
51
+ "0",
52
+ "t",
53
+ "t"
54
+ ],
55
+ [
56
+ "t",
57
+ "3",
58
+ "t",
59
+ "0",
60
+ "t"
61
+ ],
62
+ [
63
+ "t",
64
+ "1",
65
+ "t",
66
+ "t",
67
+ "0"
68
+ ]
69
+ ],
70
+ "symmetry_groups": [
71
+ [
72
+ 3
73
+ ],
74
+ [
75
+ 5
76
+ ],
77
+ [
78
+ 2,
79
+ 4,
80
+ 6
81
+ ]
82
+ ],
83
+ "orthonormal_quotient": [
84
+ [
85
+ "0",
86
+ "3",
87
+ "sqrt(3)"
88
+ ],
89
+ [
90
+ "3",
91
+ "0",
92
+ "sqrt(3)*t"
93
+ ],
94
+ [
95
+ "sqrt(3)",
96
+ "sqrt(3)*t",
97
+ "2*t"
98
+ ]
99
+ ],
100
+ "quotient_determinant": "0",
101
+ "quotient_sum_principal_2_minors": "-3*(t**2 + 4)",
102
+ "quotient_sign_certificate": "determinant=0 and e2=-3t^2-12<0; the two nonzero eigenvalues have opposite signs",
103
+ "orthogonal_modes": [
104
+ {
105
+ "eigenvalue": "-t",
106
+ "multiplicity": 2
107
+ }
108
+ ],
109
+ "reduced_hessian_inertia_positive_negative_zero": [
110
+ 1,
111
+ 3,
112
+ 1
113
+ ],
114
+ "arithmetic": "exact symbolic identities; elementary sign certificate for t>1"
115
+ },
116
+ {
117
+ "derivative_label": 2,
118
+ "remaining_labels": [
119
+ 1,
120
+ 3,
121
+ 4,
122
+ 5,
123
+ 6
124
+ ],
125
+ "hessian": [
126
+ [
127
+ "0",
128
+ "1",
129
+ "t",
130
+ "t",
131
+ "t"
132
+ ],
133
+ [
134
+ "1",
135
+ "0",
136
+ "t",
137
+ "t",
138
+ "t"
139
+ ],
140
+ [
141
+ "t",
142
+ "t",
143
+ "0",
144
+ "1",
145
+ "(2*t - 1)*(2*t + 1)/t**2"
146
+ ],
147
+ [
148
+ "t",
149
+ "t",
150
+ "1",
151
+ "0",
152
+ "1"
153
+ ],
154
+ [
155
+ "t",
156
+ "t",
157
+ "(2*t - 1)*(2*t + 1)/t**2",
158
+ "1",
159
+ "0"
160
+ ]
161
+ ],
162
+ "symmetry_groups": [
163
+ [
164
+ 1,
165
+ 3
166
+ ],
167
+ [
168
+ 4,
169
+ 6
170
+ ],
171
+ [
172
+ 5
173
+ ]
174
+ ],
175
+ "orthonormal_quotient": [
176
+ [
177
+ "1",
178
+ "2*t",
179
+ "sqrt(2)*t"
180
+ ],
181
+ [
182
+ "2*t",
183
+ "(2*t - 1)*(2*t + 1)/t**2",
184
+ "sqrt(2)"
185
+ ],
186
+ [
187
+ "sqrt(2)*t",
188
+ "sqrt(2)",
189
+ "0"
190
+ ]
191
+ ],
192
+ "quotient_determinant": "0",
193
+ "quotient_sum_principal_2_minors": "-(6*t**4 - 2*t**2 + 1)/t**2",
194
+ "quotient_sign_certificate": "determinant=0; -t^2 e2=6(t^2-1)^2+10(t^2-1)+5>0 for t>1",
195
+ "orthogonal_modes": [
196
+ {
197
+ "eigenvalue": "-1",
198
+ "multiplicity": 1
199
+ },
200
+ {
201
+ "eigenvalue": "-4 + t**(-2)",
202
+ "multiplicity": 1
203
+ }
204
+ ],
205
+ "reduced_hessian_inertia_positive_negative_zero": [
206
+ 1,
207
+ 3,
208
+ 1
209
+ ],
210
+ "arithmetic": "exact symbolic identities; elementary sign certificate for t>1"
211
+ },
212
+ {
213
+ "derivative_label": 3,
214
+ "remaining_labels": [
215
+ 1,
216
+ 2,
217
+ 4,
218
+ 5,
219
+ 6
220
+ ],
221
+ "hessian": [
222
+ [
223
+ "0",
224
+ "1",
225
+ "1",
226
+ "3",
227
+ "1"
228
+ ],
229
+ [
230
+ "1",
231
+ "0",
232
+ "t",
233
+ "t",
234
+ "t"
235
+ ],
236
+ [
237
+ "1",
238
+ "t",
239
+ "0",
240
+ "t",
241
+ "t"
242
+ ],
243
+ [
244
+ "3",
245
+ "t",
246
+ "t",
247
+ "0",
248
+ "t"
249
+ ],
250
+ [
251
+ "1",
252
+ "t",
253
+ "t",
254
+ "t",
255
+ "0"
256
+ ]
257
+ ],
258
+ "symmetry_groups": [
259
+ [
260
+ 1
261
+ ],
262
+ [
263
+ 5
264
+ ],
265
+ [
266
+ 2,
267
+ 4,
268
+ 6
269
+ ]
270
+ ],
271
+ "orthonormal_quotient": [
272
+ [
273
+ "0",
274
+ "3",
275
+ "sqrt(3)"
276
+ ],
277
+ [
278
+ "3",
279
+ "0",
280
+ "sqrt(3)*t"
281
+ ],
282
+ [
283
+ "sqrt(3)",
284
+ "sqrt(3)*t",
285
+ "2*t"
286
+ ]
287
+ ],
288
+ "quotient_determinant": "0",
289
+ "quotient_sum_principal_2_minors": "-3*(t**2 + 4)",
290
+ "quotient_sign_certificate": "determinant=0 and e2=-3t^2-12<0; the two nonzero eigenvalues have opposite signs",
291
+ "orthogonal_modes": [
292
+ {
293
+ "eigenvalue": "-t",
294
+ "multiplicity": 2
295
+ }
296
+ ],
297
+ "reduced_hessian_inertia_positive_negative_zero": [
298
+ 1,
299
+ 3,
300
+ 1
301
+ ],
302
+ "arithmetic": "exact symbolic identities; elementary sign certificate for t>1"
303
+ },
304
+ {
305
+ "derivative_label": 4,
306
+ "remaining_labels": [
307
+ 1,
308
+ 2,
309
+ 3,
310
+ 5,
311
+ 6
312
+ ],
313
+ "hessian": [
314
+ [
315
+ "0",
316
+ "t",
317
+ "1",
318
+ "t",
319
+ "t"
320
+ ],
321
+ [
322
+ "t",
323
+ "0",
324
+ "t",
325
+ "1",
326
+ "(2*t - 1)*(2*t + 1)/t**2"
327
+ ],
328
+ [
329
+ "1",
330
+ "t",
331
+ "0",
332
+ "t",
333
+ "t"
334
+ ],
335
+ [
336
+ "t",
337
+ "1",
338
+ "t",
339
+ "0",
340
+ "1"
341
+ ],
342
+ [
343
+ "t",
344
+ "(2*t - 1)*(2*t + 1)/t**2",
345
+ "t",
346
+ "1",
347
+ "0"
348
+ ]
349
+ ],
350
+ "symmetry_groups": [
351
+ [
352
+ 1,
353
+ 3
354
+ ],
355
+ [
356
+ 2,
357
+ 6
358
+ ],
359
+ [
360
+ 5
361
+ ]
362
+ ],
363
+ "orthonormal_quotient": [
364
+ [
365
+ "1",
366
+ "2*t",
367
+ "sqrt(2)*t"
368
+ ],
369
+ [
370
+ "2*t",
371
+ "(2*t - 1)*(2*t + 1)/t**2",
372
+ "sqrt(2)"
373
+ ],
374
+ [
375
+ "sqrt(2)*t",
376
+ "sqrt(2)",
377
+ "0"
378
+ ]
379
+ ],
380
+ "quotient_determinant": "0",
381
+ "quotient_sum_principal_2_minors": "-(6*t**4 - 2*t**2 + 1)/t**2",
382
+ "quotient_sign_certificate": "determinant=0; -t^2 e2=6(t^2-1)^2+10(t^2-1)+5>0 for t>1",
383
+ "orthogonal_modes": [
384
+ {
385
+ "eigenvalue": "-1",
386
+ "multiplicity": 1
387
+ },
388
+ {
389
+ "eigenvalue": "-4 + t**(-2)",
390
+ "multiplicity": 1
391
+ }
392
+ ],
393
+ "reduced_hessian_inertia_positive_negative_zero": [
394
+ 1,
395
+ 3,
396
+ 1
397
+ ],
398
+ "arithmetic": "exact symbolic identities; elementary sign certificate for t>1"
399
+ },
400
+ {
401
+ "derivative_label": 5,
402
+ "remaining_labels": [
403
+ 1,
404
+ 2,
405
+ 3,
406
+ 4,
407
+ 6
408
+ ],
409
+ "hessian": [
410
+ [
411
+ "0",
412
+ "t",
413
+ "3",
414
+ "t",
415
+ "t"
416
+ ],
417
+ [
418
+ "t",
419
+ "0",
420
+ "t",
421
+ "1",
422
+ "1"
423
+ ],
424
+ [
425
+ "3",
426
+ "t",
427
+ "0",
428
+ "t",
429
+ "t"
430
+ ],
431
+ [
432
+ "t",
433
+ "1",
434
+ "t",
435
+ "0",
436
+ "1"
437
+ ],
438
+ [
439
+ "t",
440
+ "1",
441
+ "t",
442
+ "1",
443
+ "0"
444
+ ]
445
+ ],
446
+ "symmetry_groups": [
447
+ [
448
+ 1,
449
+ 3
450
+ ],
451
+ [
452
+ 2,
453
+ 4,
454
+ 6
455
+ ]
456
+ ],
457
+ "orthonormal_quotient": [
458
+ [
459
+ "3",
460
+ "sqrt(6)*t"
461
+ ],
462
+ [
463
+ "sqrt(6)*t",
464
+ "2"
465
+ ]
466
+ ],
467
+ "quotient_determinant": "-6*(t - 1)*(t + 1)",
468
+ "quotient_sum_principal_2_minors": "-6*(t - 1)*(t + 1)",
469
+ "quotient_sign_certificate": "determinant=6(1-t^2)<0 for t>1; two eigenvalues have opposite signs",
470
+ "orthogonal_modes": [
471
+ {
472
+ "eigenvalue": "-3",
473
+ "multiplicity": 1
474
+ },
475
+ {
476
+ "eigenvalue": "-1",
477
+ "multiplicity": 2
478
+ }
479
+ ],
480
+ "reduced_hessian_inertia_positive_negative_zero": [
481
+ 1,
482
+ 4,
483
+ 0
484
+ ],
485
+ "arithmetic": "exact symbolic identities; elementary sign certificate for t>1"
486
+ },
487
+ {
488
+ "derivative_label": 6,
489
+ "remaining_labels": [
490
+ 1,
491
+ 2,
492
+ 3,
493
+ 4,
494
+ 5
495
+ ],
496
+ "hessian": [
497
+ [
498
+ "0",
499
+ "t",
500
+ "1",
501
+ "t",
502
+ "t"
503
+ ],
504
+ [
505
+ "t",
506
+ "0",
507
+ "t",
508
+ "(2*t - 1)*(2*t + 1)/t**2",
509
+ "1"
510
+ ],
511
+ [
512
+ "1",
513
+ "t",
514
+ "0",
515
+ "t",
516
+ "t"
517
+ ],
518
+ [
519
+ "t",
520
+ "(2*t - 1)*(2*t + 1)/t**2",
521
+ "t",
522
+ "0",
523
+ "1"
524
+ ],
525
+ [
526
+ "t",
527
+ "1",
528
+ "t",
529
+ "1",
530
+ "0"
531
+ ]
532
+ ],
533
+ "symmetry_groups": [
534
+ [
535
+ 1,
536
+ 3
537
+ ],
538
+ [
539
+ 2,
540
+ 4
541
+ ],
542
+ [
543
+ 5
544
+ ]
545
+ ],
546
+ "orthonormal_quotient": [
547
+ [
548
+ "1",
549
+ "2*t",
550
+ "sqrt(2)*t"
551
+ ],
552
+ [
553
+ "2*t",
554
+ "(2*t - 1)*(2*t + 1)/t**2",
555
+ "sqrt(2)"
556
+ ],
557
+ [
558
+ "sqrt(2)*t",
559
+ "sqrt(2)",
560
+ "0"
561
+ ]
562
+ ],
563
+ "quotient_determinant": "0",
564
+ "quotient_sum_principal_2_minors": "-(6*t**4 - 2*t**2 + 1)/t**2",
565
+ "quotient_sign_certificate": "determinant=0; -t^2 e2=6(t^2-1)^2+10(t^2-1)+5>0 for t>1",
566
+ "orthogonal_modes": [
567
+ {
568
+ "eigenvalue": "-1",
569
+ "multiplicity": 1
570
+ },
571
+ {
572
+ "eigenvalue": "-4 + t**(-2)",
573
+ "multiplicity": 1
574
+ }
575
+ ],
576
+ "reduced_hessian_inertia_positive_negative_zero": [
577
+ 1,
578
+ 3,
579
+ 1
580
+ ],
581
+ "arithmetic": "exact symbolic identities; elementary sign certificate for t>1"
582
+ }
583
+ ],
584
+ "strong_relation": {
585
+ "I": [
586
+ 1,
587
+ 3
588
+ ],
589
+ "J": [
590
+ 2,
591
+ 4,
592
+ 5,
593
+ 6
594
+ ],
595
+ "products": [
596
+ "1",
597
+ "1",
598
+ "3*(2*t - 1)*(2*t + 1)/t**2",
599
+ "1"
600
+ ],
601
+ "necessary_parameter": "p >= log_3(12-3/t^2)",
602
+ "uniform_lower_bound": "p >= log_3(12)"
603
+ },
604
+ "integer_example": {
605
+ "parameter_t": 2,
606
+ "global_coefficient_scale": 4,
607
+ "coefficients": {
608
+ "123": 4,
609
+ "124": 8,
610
+ "125": 8,
611
+ "126": 8,
612
+ "134": 4,
613
+ "135": 12,
614
+ "136": 4,
615
+ "145": 8,
616
+ "146": 8,
617
+ "156": 8,
618
+ "234": 8,
619
+ "235": 8,
620
+ "236": 8,
621
+ "245": 4,
622
+ "246": 15,
623
+ "256": 4,
624
+ "345": 8,
625
+ "346": 8,
626
+ "356": 8,
627
+ "456": 4
628
+ },
629
+ "strong_relation_products": [
630
+ 16,
631
+ 16,
632
+ 180,
633
+ 16
634
+ ],
635
+ "exact_squared_T2_excess": 36,
636
+ "rational_principal_minors_checked": 186
637
+ },
638
+ "strict_interior_family": {
639
+ "domain": "t>1 and 0<delta<1",
640
+ "coefficients": {
641
+ "PPs": "3-delta",
642
+ "QQQ": "4-t^(-2)-delta",
643
+ "PPQ": "1",
644
+ "sQQ": "1",
645
+ "PsQ": "t",
646
+ "PQQ": "t"
647
+ },
648
+ "derivative_certificates": [
649
+ {
650
+ "derivative_label": 1,
651
+ "orthonormal_quotient": [
652
+ [
653
+ "0",
654
+ "3 - delta",
655
+ "sqrt(3)"
656
+ ],
657
+ [
658
+ "3 - delta",
659
+ "0",
660
+ "sqrt(3)*t"
661
+ ],
662
+ [
663
+ "sqrt(3)",
664
+ "sqrt(3)*t",
665
+ "2*t"
666
+ ]
667
+ ],
668
+ "quotient_determinant": "-2*delta*t*(delta - 3)",
669
+ "quotient_sum_principal_2_minors": "-delta**2 + 6*delta - 3*t**2 - 12",
670
+ "strict_sign_certificate": "determinant=2(3-delta)t delta>0 and e2=-(3-delta)^2-3-3t^2<0; one positive and two negative quotient eigenvalues",
671
+ "orthogonal_modes": [
672
+ {
673
+ "eigenvalue": "-t",
674
+ "multiplicity": 2
675
+ }
676
+ ],
677
+ "reduced_hessian_inertia_positive_negative_zero": [
678
+ 1,
679
+ 4,
680
+ 0
681
+ ]
682
+ },
683
+ {
684
+ "derivative_label": 2,
685
+ "orthonormal_quotient": [
686
+ [
687
+ "1",
688
+ "2*t",
689
+ "sqrt(2)*t"
690
+ ],
691
+ [
692
+ "2*t",
693
+ "-(delta*t**2 - 4*t**2 + 1)/t**2",
694
+ "sqrt(2)"
695
+ ],
696
+ [
697
+ "sqrt(2)*t",
698
+ "sqrt(2)",
699
+ "0"
700
+ ]
701
+ ],
702
+ "quotient_determinant": "2*delta*t**2",
703
+ "quotient_sum_principal_2_minors": "-(delta*t**2 + 6*t**4 - 2*t**2 + 1)/t**2",
704
+ "strict_sign_certificate": "determinant=2t^2delta>0 and e2=2-t^(-2)-delta-6t^2<0; one positive and two negative quotient eigenvalues",
705
+ "orthogonal_modes": [
706
+ {
707
+ "eigenvalue": "-1",
708
+ "multiplicity": 1
709
+ },
710
+ {
711
+ "eigenvalue": "delta-4+t^(-2)",
712
+ "multiplicity": 1
713
+ }
714
+ ],
715
+ "reduced_hessian_inertia_positive_negative_zero": [
716
+ 1,
717
+ 4,
718
+ 0
719
+ ]
720
+ },
721
+ {
722
+ "derivative_label": 3,
723
+ "orthonormal_quotient": [
724
+ [
725
+ "0",
726
+ "3 - delta",
727
+ "sqrt(3)"
728
+ ],
729
+ [
730
+ "3 - delta",
731
+ "0",
732
+ "sqrt(3)*t"
733
+ ],
734
+ [
735
+ "sqrt(3)",
736
+ "sqrt(3)*t",
737
+ "2*t"
738
+ ]
739
+ ],
740
+ "quotient_determinant": "-2*delta*t*(delta - 3)",
741
+ "quotient_sum_principal_2_minors": "-delta**2 + 6*delta - 3*t**2 - 12",
742
+ "strict_sign_certificate": "determinant=2(3-delta)t delta>0 and e2=-(3-delta)^2-3-3t^2<0; one positive and two negative quotient eigenvalues",
743
+ "orthogonal_modes": [
744
+ {
745
+ "eigenvalue": "-t",
746
+ "multiplicity": 2
747
+ }
748
+ ],
749
+ "reduced_hessian_inertia_positive_negative_zero": [
750
+ 1,
751
+ 4,
752
+ 0
753
+ ]
754
+ },
755
+ {
756
+ "derivative_label": 4,
757
+ "orthonormal_quotient": [
758
+ [
759
+ "1",
760
+ "2*t",
761
+ "sqrt(2)*t"
762
+ ],
763
+ [
764
+ "2*t",
765
+ "-(delta*t**2 - 4*t**2 + 1)/t**2",
766
+ "sqrt(2)"
767
+ ],
768
+ [
769
+ "sqrt(2)*t",
770
+ "sqrt(2)",
771
+ "0"
772
+ ]
773
+ ],
774
+ "quotient_determinant": "2*delta*t**2",
775
+ "quotient_sum_principal_2_minors": "-(delta*t**2 + 6*t**4 - 2*t**2 + 1)/t**2",
776
+ "strict_sign_certificate": "determinant=2t^2delta>0 and e2=2-t^(-2)-delta-6t^2<0; one positive and two negative quotient eigenvalues",
777
+ "orthogonal_modes": [
778
+ {
779
+ "eigenvalue": "-1",
780
+ "multiplicity": 1
781
+ },
782
+ {
783
+ "eigenvalue": "delta-4+t^(-2)",
784
+ "multiplicity": 1
785
+ }
786
+ ],
787
+ "reduced_hessian_inertia_positive_negative_zero": [
788
+ 1,
789
+ 4,
790
+ 0
791
+ ]
792
+ },
793
+ {
794
+ "derivative_label": 5,
795
+ "orthonormal_quotient": [
796
+ [
797
+ "3 - delta",
798
+ "sqrt(6)*t"
799
+ ],
800
+ [
801
+ "sqrt(6)*t",
802
+ "2"
803
+ ]
804
+ ],
805
+ "quotient_determinant": "-2*(delta + 3*t**2 - 3)",
806
+ "quotient_sum_principal_2_minors": "-2*(delta + 3*t**2 - 3)",
807
+ "strict_sign_certificate": "determinant=-6(t^2-1)-2delta<0; two quotient eigenvalues have opposite signs",
808
+ "orthogonal_modes": [
809
+ {
810
+ "eigenvalue": "delta-3",
811
+ "multiplicity": 1
812
+ },
813
+ {
814
+ "eigenvalue": "-1",
815
+ "multiplicity": 2
816
+ }
817
+ ],
818
+ "reduced_hessian_inertia_positive_negative_zero": [
819
+ 1,
820
+ 4,
821
+ 0
822
+ ]
823
+ },
824
+ {
825
+ "derivative_label": 6,
826
+ "orthonormal_quotient": [
827
+ [
828
+ "1",
829
+ "2*t",
830
+ "sqrt(2)*t"
831
+ ],
832
+ [
833
+ "2*t",
834
+ "-(delta*t**2 - 4*t**2 + 1)/t**2",
835
+ "sqrt(2)"
836
+ ],
837
+ [
838
+ "sqrt(2)*t",
839
+ "sqrt(2)",
840
+ "0"
841
+ ]
842
+ ],
843
+ "quotient_determinant": "2*delta*t**2",
844
+ "quotient_sum_principal_2_minors": "-(delta*t**2 + 6*t**4 - 2*t**2 + 1)/t**2",
845
+ "strict_sign_certificate": "determinant=2t^2delta>0 and e2=2-t^(-2)-delta-6t^2<0; one positive and two negative quotient eigenvalues",
846
+ "orthogonal_modes": [
847
+ {
848
+ "eigenvalue": "-1",
849
+ "multiplicity": 1
850
+ },
851
+ {
852
+ "eigenvalue": "delta-4+t^(-2)",
853
+ "multiplicity": 1
854
+ }
855
+ ],
856
+ "reduced_hessian_inertia_positive_negative_zero": [
857
+ 1,
858
+ 4,
859
+ 0
860
+ ]
861
+ }
862
+ ],
863
+ "strong_relation_products": [
864
+ "1",
865
+ "1",
866
+ "(delta - 3)*(delta*t**2 - 4*t**2 + 1)/t**2",
867
+ "1"
868
+ ],
869
+ "necessary_parameter": "p >= log_3((3-delta)(4-t^(-2)-delta)) whenever the product exceeds 1",
870
+ "limiting_uniform_lower_bound": "p >= log_3(12)",
871
+ "strict_integer_example": {
872
+ "parameter_t": 2,
873
+ "parameter_delta": "1/4",
874
+ "global_coefficient_scale": 4,
875
+ "coefficients": {
876
+ "123": 4,
877
+ "124": 8,
878
+ "125": 8,
879
+ "126": 8,
880
+ "134": 4,
881
+ "135": 11,
882
+ "136": 4,
883
+ "145": 8,
884
+ "146": 8,
885
+ "156": 8,
886
+ "234": 8,
887
+ "235": 8,
888
+ "236": 8,
889
+ "245": 4,
890
+ "246": 14,
891
+ "256": 4,
892
+ "345": 8,
893
+ "346": 8,
894
+ "356": 8,
895
+ "456": 4
896
+ },
897
+ "strong_relation_products": [
898
+ 16,
899
+ 16,
900
+ 154,
901
+ 16
902
+ ],
903
+ "exact_squared_T2_excess": 10,
904
+ "rational_principal_minors_checked": 186,
905
+ "qualification": "All six reduced derivative Hessians have signature (1,4).",
906
+ "explicit_absolute_coefficient_perturbation_radius": "1/1000",
907
+ "hessian_determinants": [
908
+ 11264,
909
+ 7168,
910
+ 11264,
911
+ 7168,
912
+ 52096,
913
+ 7168
914
+ ],
915
+ "hessian_maximum_absolute_row_sums": [
916
+ 35,
917
+ 34,
918
+ 35,
919
+ 34,
920
+ 35,
921
+ 34
922
+ ],
923
+ "minimum_eigenvalue_absolute_lower_bound": "1024/214375",
924
+ "perturbation_operator_norm_upper_bound": "1/250",
925
+ "worst_case_squared_T2_failure_margin": "618937/62500"
926
+ }
927
+ },
928
+ "scope": "This lower bound does not establish a matching upper bound or resolve finiteness of the strong parameter.",
929
+ "status": "all exact symbolic and rational checks passed"
930
+ }
results/strengthened_verification_results.json ADDED
@@ -0,0 +1,2109 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "title": "Strengthened aggregation reproducibility certificates",
3
+ "scope": "Finite checks supplement the self-contained mathematical proofs.",
4
+ "exact_quadratic_congruence": [
5
+ {
6
+ "trial": 1,
7
+ "input_parameter": 1,
8
+ "source_check": {
9
+ "quartet_count": 126,
10
+ "minimum_exact_polynomial_margin": "0"
11
+ },
12
+ "first_substitution": {
13
+ "quartet_count": 35,
14
+ "minimum_exact_polynomial_margin": "0"
15
+ },
16
+ "composed_substitution": {
17
+ "quartet_count": 15,
18
+ "minimum_exact_polynomial_margin": "14161/4"
19
+ },
20
+ "nonzero_diagonal_after_first": 3,
21
+ "exact_composition_identity": true,
22
+ "arithmetic": "exact rational"
23
+ },
24
+ {
25
+ "trial": 1,
26
+ "input_parameter": 2,
27
+ "source_check": {
28
+ "quartet_count": 126,
29
+ "minimum_exact_polynomial_margin": "0"
30
+ },
31
+ "first_substitution": {
32
+ "quartet_count": 35,
33
+ "minimum_exact_polynomial_margin": "0"
34
+ },
35
+ "composed_substitution": {
36
+ "quartet_count": 15,
37
+ "minimum_exact_polynomial_margin": "601601763/16"
38
+ },
39
+ "nonzero_diagonal_after_first": 3,
40
+ "exact_composition_identity": true,
41
+ "arithmetic": "exact rational"
42
+ },
43
+ {
44
+ "trial": 2,
45
+ "input_parameter": 1,
46
+ "source_check": {
47
+ "quartet_count": 126,
48
+ "minimum_exact_polynomial_margin": "0"
49
+ },
50
+ "first_substitution": {
51
+ "quartet_count": 35,
52
+ "minimum_exact_polynomial_margin": "9604/81"
53
+ },
54
+ "composed_substitution": {
55
+ "quartet_count": 15,
56
+ "minimum_exact_polynomial_margin": "11703241/81"
57
+ },
58
+ "nonzero_diagonal_after_first": 4,
59
+ "exact_composition_identity": true,
60
+ "arithmetic": "exact rational"
61
+ },
62
+ {
63
+ "trial": 2,
64
+ "input_parameter": 2,
65
+ "source_check": {
66
+ "quartet_count": 126,
67
+ "minimum_exact_polynomial_margin": "0"
68
+ },
69
+ "first_substitution": {
70
+ "quartet_count": 35,
71
+ "minimum_exact_polynomial_margin": "21293813776/2187"
72
+ },
73
+ "composed_substitution": {
74
+ "quartet_count": 15,
75
+ "minimum_exact_polynomial_margin": "2539736758431601/27"
76
+ },
77
+ "nonzero_diagonal_after_first": 4,
78
+ "exact_composition_identity": true,
79
+ "arithmetic": "exact rational"
80
+ },
81
+ {
82
+ "trial": 3,
83
+ "input_parameter": 1,
84
+ "source_check": {
85
+ "quartet_count": 126,
86
+ "minimum_exact_polynomial_margin": "0"
87
+ },
88
+ "first_substitution": {
89
+ "quartet_count": 35,
90
+ "minimum_exact_polynomial_margin": "64"
91
+ },
92
+ "composed_substitution": {
93
+ "quartet_count": 15,
94
+ "minimum_exact_polynomial_margin": "360000"
95
+ },
96
+ "nonzero_diagonal_after_first": 4,
97
+ "exact_composition_identity": true,
98
+ "arithmetic": "exact rational"
99
+ },
100
+ {
101
+ "trial": 3,
102
+ "input_parameter": 2,
103
+ "source_check": {
104
+ "quartet_count": 126,
105
+ "minimum_exact_polynomial_margin": "0"
106
+ },
107
+ "first_substitution": {
108
+ "quartet_count": 35,
109
+ "minimum_exact_polynomial_margin": "14776336/27"
110
+ },
111
+ "composed_substitution": {
112
+ "quartet_count": 15,
113
+ "minimum_exact_polynomial_margin": "3273452781097216/27"
114
+ },
115
+ "nonzero_diagonal_after_first": 4,
116
+ "exact_composition_identity": true,
117
+ "arithmetic": "exact rational"
118
+ },
119
+ {
120
+ "trial": 4,
121
+ "input_parameter": 1,
122
+ "source_check": {
123
+ "quartet_count": 126,
124
+ "minimum_exact_polynomial_margin": "0"
125
+ },
126
+ "first_substitution": {
127
+ "quartet_count": 35,
128
+ "minimum_exact_polynomial_margin": "64/9"
129
+ },
130
+ "composed_substitution": {
131
+ "quartet_count": 15,
132
+ "minimum_exact_polynomial_margin": "784"
133
+ },
134
+ "nonzero_diagonal_after_first": 4,
135
+ "exact_composition_identity": true,
136
+ "arithmetic": "exact rational"
137
+ },
138
+ {
139
+ "trial": 4,
140
+ "input_parameter": 2,
141
+ "source_check": {
142
+ "quartet_count": 126,
143
+ "minimum_exact_polynomial_margin": "0"
144
+ },
145
+ "first_substitution": {
146
+ "quartet_count": 35,
147
+ "minimum_exact_polynomial_margin": "4096/27"
148
+ },
149
+ "composed_substitution": {
150
+ "quartet_count": 15,
151
+ "minimum_exact_polynomial_margin": "7680000"
152
+ },
153
+ "nonzero_diagonal_after_first": 4,
154
+ "exact_composition_identity": true,
155
+ "arithmetic": "exact rational"
156
+ },
157
+ {
158
+ "trial": 5,
159
+ "input_parameter": 1,
160
+ "source_check": {
161
+ "quartet_count": 126,
162
+ "minimum_exact_polynomial_margin": "0"
163
+ },
164
+ "first_substitution": {
165
+ "quartet_count": 35,
166
+ "minimum_exact_polynomial_margin": "400"
167
+ },
168
+ "composed_substitution": {
169
+ "quartet_count": 15,
170
+ "minimum_exact_polynomial_margin": "1800964"
171
+ },
172
+ "nonzero_diagonal_after_first": 4,
173
+ "exact_composition_identity": true,
174
+ "arithmetic": "exact rational"
175
+ },
176
+ {
177
+ "trial": 5,
178
+ "input_parameter": 2,
179
+ "source_check": {
180
+ "quartet_count": 126,
181
+ "minimum_exact_polynomial_margin": "0"
182
+ },
183
+ "first_substitution": {
184
+ "quartet_count": 35,
185
+ "minimum_exact_polynomial_margin": "236421376/27"
186
+ },
187
+ "composed_substitution": {
188
+ "quartet_count": 15,
189
+ "minimum_exact_polynomial_margin": "19946755118430000"
190
+ },
191
+ "nonzero_diagonal_after_first": 4,
192
+ "exact_composition_identity": true,
193
+ "arithmetic": "exact rational"
194
+ },
195
+ {
196
+ "trial": 6,
197
+ "input_parameter": 1,
198
+ "source_check": {
199
+ "quartet_count": 126,
200
+ "minimum_exact_polynomial_margin": "0"
201
+ },
202
+ "first_substitution": {
203
+ "quartet_count": 35,
204
+ "minimum_exact_polynomial_margin": "400/81"
205
+ },
206
+ "composed_substitution": {
207
+ "quartet_count": 15,
208
+ "minimum_exact_polynomial_margin": "5368489/81"
209
+ },
210
+ "nonzero_diagonal_after_first": 4,
211
+ "exact_composition_identity": true,
212
+ "arithmetic": "exact rational"
213
+ },
214
+ {
215
+ "trial": 6,
216
+ "input_parameter": 2,
217
+ "source_check": {
218
+ "quartet_count": 126,
219
+ "minimum_exact_polynomial_margin": "0"
220
+ },
221
+ "first_substitution": {
222
+ "quartet_count": 35,
223
+ "minimum_exact_polynomial_margin": "100000000/2187"
224
+ },
225
+ "composed_substitution": {
226
+ "quartet_count": 15,
227
+ "minimum_exact_polynomial_margin": "84290069450005441/6912"
228
+ },
229
+ "nonzero_diagonal_after_first": 4,
230
+ "exact_composition_identity": true,
231
+ "arithmetic": "exact rational"
232
+ },
233
+ {
234
+ "trial": 7,
235
+ "input_parameter": 1,
236
+ "source_check": {
237
+ "quartet_count": 126,
238
+ "minimum_exact_polynomial_margin": "0"
239
+ },
240
+ "first_substitution": {
241
+ "quartet_count": 35,
242
+ "minimum_exact_polynomial_margin": "0"
243
+ },
244
+ "composed_substitution": {
245
+ "quartet_count": 15,
246
+ "minimum_exact_polynomial_margin": "1156/9"
247
+ },
248
+ "nonzero_diagonal_after_first": 3,
249
+ "exact_composition_identity": true,
250
+ "arithmetic": "exact rational"
251
+ },
252
+ {
253
+ "trial": 7,
254
+ "input_parameter": 2,
255
+ "source_check": {
256
+ "quartet_count": 126,
257
+ "minimum_exact_polynomial_margin": "0"
258
+ },
259
+ "first_substitution": {
260
+ "quartet_count": 35,
261
+ "minimum_exact_polynomial_margin": "0"
262
+ },
263
+ "composed_substitution": {
264
+ "quartet_count": 15,
265
+ "minimum_exact_polynomial_margin": "2560000/27"
266
+ },
267
+ "nonzero_diagonal_after_first": 3,
268
+ "exact_composition_identity": true,
269
+ "arithmetic": "exact rational"
270
+ },
271
+ {
272
+ "trial": 8,
273
+ "input_parameter": 1,
274
+ "source_check": {
275
+ "quartet_count": 126,
276
+ "minimum_exact_polynomial_margin": "0"
277
+ },
278
+ "first_substitution": {
279
+ "quartet_count": 35,
280
+ "minimum_exact_polynomial_margin": "9"
281
+ },
282
+ "composed_substitution": {
283
+ "quartet_count": 15,
284
+ "minimum_exact_polynomial_margin": "70879561/324"
285
+ },
286
+ "nonzero_diagonal_after_first": 4,
287
+ "exact_composition_identity": true,
288
+ "arithmetic": "exact rational"
289
+ },
290
+ {
291
+ "trial": 8,
292
+ "input_parameter": 2,
293
+ "source_check": {
294
+ "quartet_count": 126,
295
+ "minimum_exact_polynomial_margin": "0"
296
+ },
297
+ "first_substitution": {
298
+ "quartet_count": 35,
299
+ "minimum_exact_polynomial_margin": "243"
300
+ },
301
+ "composed_substitution": {
302
+ "quartet_count": 15,
303
+ "minimum_exact_polynomial_margin": "772335182217300625/34992"
304
+ },
305
+ "nonzero_diagonal_after_first": 4,
306
+ "exact_composition_identity": true,
307
+ "arithmetic": "exact rational"
308
+ }
309
+ ],
310
+ "formal_tensor_factorizations": {
311
+ "arithmetic": "exact commutative polynomial identities",
312
+ "source_indices": 3,
313
+ "target_indices": 4,
314
+ "factorizations": [
315
+ {
316
+ "pairing": [
317
+ [
318
+ 0,
319
+ 1
320
+ ],
321
+ [
322
+ 2,
323
+ 3
324
+ ]
325
+ ],
326
+ "expanded_monomials": 81,
327
+ "total_summands": 81,
328
+ "status": "identity verified"
329
+ },
330
+ {
331
+ "pairing": [
332
+ [
333
+ 0,
334
+ 2
335
+ ],
336
+ [
337
+ 1,
338
+ 3
339
+ ]
340
+ ],
341
+ "expanded_monomials": 81,
342
+ "total_summands": 81,
343
+ "status": "identity verified"
344
+ },
345
+ {
346
+ "pairing": [
347
+ [
348
+ 0,
349
+ 3
350
+ ],
351
+ [
352
+ 1,
353
+ 2
354
+ ]
355
+ ],
356
+ "expanded_monomials": 81,
357
+ "total_summands": 81,
358
+ "status": "identity verified"
359
+ }
360
+ ]
361
+ },
362
+ "numerical_quadratic_congruence": [
363
+ {
364
+ "trial": 1,
365
+ "input_parameter": "0.25",
366
+ "output_parameter": "1.0",
367
+ "stage": "first",
368
+ "quartet_count": 35,
369
+ "minimum_relative_root_polygon_slack": "0.0",
370
+ "precision_decimal_digits": 90,
371
+ "arithmetic": "high-precision numerical; not an exact certificate"
372
+ },
373
+ {
374
+ "trial": 1,
375
+ "input_parameter": "0.25",
376
+ "output_parameter": "1.0",
377
+ "stage": "composed",
378
+ "quartet_count": 15,
379
+ "minimum_relative_root_polygon_slack": "0.84697361048804553039197642515",
380
+ "precision_decimal_digits": 90,
381
+ "arithmetic": "high-precision numerical; not an exact certificate"
382
+ },
383
+ {
384
+ "trial": 1,
385
+ "input_parameter": "0.5",
386
+ "output_parameter": "1.0",
387
+ "stage": "first",
388
+ "quartet_count": 35,
389
+ "minimum_relative_root_polygon_slack": "0.0",
390
+ "precision_decimal_digits": 90,
391
+ "arithmetic": "high-precision numerical; not an exact certificate"
392
+ },
393
+ {
394
+ "trial": 1,
395
+ "input_parameter": "0.5",
396
+ "output_parameter": "1.0",
397
+ "stage": "composed",
398
+ "quartet_count": 15,
399
+ "minimum_relative_root_polygon_slack": "0.830270557766429988210651990218",
400
+ "precision_decimal_digits": 90,
401
+ "arithmetic": "high-precision numerical; not an exact certificate"
402
+ },
403
+ {
404
+ "trial": 1,
405
+ "input_parameter": "1",
406
+ "output_parameter": "1.0",
407
+ "stage": "first",
408
+ "quartet_count": 35,
409
+ "minimum_relative_root_polygon_slack": "0.0",
410
+ "precision_decimal_digits": 90,
411
+ "arithmetic": "high-precision numerical; not an exact certificate"
412
+ },
413
+ {
414
+ "trial": 1,
415
+ "input_parameter": "1",
416
+ "output_parameter": "1.0",
417
+ "stage": "composed",
418
+ "quartet_count": 15,
419
+ "minimum_relative_root_polygon_slack": "0.786204193611601019008426415834",
420
+ "precision_decimal_digits": 90,
421
+ "arithmetic": "high-precision numerical; not an exact certificate"
422
+ },
423
+ {
424
+ "trial": 1,
425
+ "input_parameter": "1.5",
426
+ "output_parameter": "1.5",
427
+ "stage": "first",
428
+ "quartet_count": 35,
429
+ "minimum_relative_root_polygon_slack": "0.0",
430
+ "precision_decimal_digits": 90,
431
+ "arithmetic": "high-precision numerical; not an exact certificate"
432
+ },
433
+ {
434
+ "trial": 1,
435
+ "input_parameter": "1.5",
436
+ "output_parameter": "1.5",
437
+ "stage": "composed",
438
+ "quartet_count": 15,
439
+ "minimum_relative_root_polygon_slack": "0.808102596544018626343781379637",
440
+ "precision_decimal_digits": 90,
441
+ "arithmetic": "high-precision numerical; not an exact certificate"
442
+ },
443
+ {
444
+ "trial": 1,
445
+ "input_parameter": "2",
446
+ "output_parameter": "2.0",
447
+ "stage": "first",
448
+ "quartet_count": 35,
449
+ "minimum_relative_root_polygon_slack": "0.0",
450
+ "precision_decimal_digits": 90,
451
+ "arithmetic": "high-precision numerical; not an exact certificate"
452
+ },
453
+ {
454
+ "trial": 1,
455
+ "input_parameter": "2",
456
+ "output_parameter": "2.0",
457
+ "stage": "composed",
458
+ "quartet_count": 15,
459
+ "minimum_relative_root_polygon_slack": "0.807157523643220592107765107997",
460
+ "precision_decimal_digits": 90,
461
+ "arithmetic": "high-precision numerical; not an exact certificate"
462
+ },
463
+ {
464
+ "trial": 1,
465
+ "input_parameter": "3",
466
+ "output_parameter": "3.0",
467
+ "stage": "first",
468
+ "quartet_count": 35,
469
+ "minimum_relative_root_polygon_slack": "0.0",
470
+ "precision_decimal_digits": 90,
471
+ "arithmetic": "high-precision numerical; not an exact certificate"
472
+ },
473
+ {
474
+ "trial": 1,
475
+ "input_parameter": "3",
476
+ "output_parameter": "3.0",
477
+ "stage": "composed",
478
+ "quartet_count": 15,
479
+ "minimum_relative_root_polygon_slack": "0.779453892787667178940834198707",
480
+ "precision_decimal_digits": 90,
481
+ "arithmetic": "high-precision numerical; not an exact certificate"
482
+ },
483
+ {
484
+ "trial": 1,
485
+ "input_parameter": "4",
486
+ "output_parameter": "4.0",
487
+ "stage": "first",
488
+ "quartet_count": 35,
489
+ "minimum_relative_root_polygon_slack": "0.0",
490
+ "precision_decimal_digits": 90,
491
+ "arithmetic": "high-precision numerical; not an exact certificate"
492
+ },
493
+ {
494
+ "trial": 1,
495
+ "input_parameter": "4",
496
+ "output_parameter": "4.0",
497
+ "stage": "composed",
498
+ "quartet_count": 15,
499
+ "minimum_relative_root_polygon_slack": "0.741394421445311092432997378321",
500
+ "precision_decimal_digits": 90,
501
+ "arithmetic": "high-precision numerical; not an exact certificate"
502
+ },
503
+ {
504
+ "trial": 1,
505
+ "input_parameter": "8",
506
+ "output_parameter": "8.0",
507
+ "stage": "first",
508
+ "quartet_count": 35,
509
+ "minimum_relative_root_polygon_slack": "0.0",
510
+ "precision_decimal_digits": 90,
511
+ "arithmetic": "high-precision numerical; not an exact certificate"
512
+ },
513
+ {
514
+ "trial": 1,
515
+ "input_parameter": "8",
516
+ "output_parameter": "8.0",
517
+ "stage": "composed",
518
+ "quartet_count": 15,
519
+ "minimum_relative_root_polygon_slack": "0.630801673063612277722412112573",
520
+ "precision_decimal_digits": 90,
521
+ "arithmetic": "high-precision numerical; not an exact certificate"
522
+ },
523
+ {
524
+ "trial": 2,
525
+ "input_parameter": "0.25",
526
+ "output_parameter": "1.0",
527
+ "stage": "first",
528
+ "quartet_count": 35,
529
+ "minimum_relative_root_polygon_slack": "0.56142028914552952847280295269",
530
+ "precision_decimal_digits": 90,
531
+ "arithmetic": "high-precision numerical; not an exact certificate"
532
+ },
533
+ {
534
+ "trial": 2,
535
+ "input_parameter": "0.25",
536
+ "output_parameter": "1.0",
537
+ "stage": "composed",
538
+ "quartet_count": 15,
539
+ "minimum_relative_root_polygon_slack": "0.978590193318809793014901820256",
540
+ "precision_decimal_digits": 90,
541
+ "arithmetic": "high-precision numerical; not an exact certificate"
542
+ },
543
+ {
544
+ "trial": 2,
545
+ "input_parameter": "0.5",
546
+ "output_parameter": "1.0",
547
+ "stage": "first",
548
+ "quartet_count": 35,
549
+ "minimum_relative_root_polygon_slack": "0.697762862487139829375436710848",
550
+ "precision_decimal_digits": 90,
551
+ "arithmetic": "high-precision numerical; not an exact certificate"
552
+ },
553
+ {
554
+ "trial": 2,
555
+ "input_parameter": "0.5",
556
+ "output_parameter": "1.0",
557
+ "stage": "composed",
558
+ "quartet_count": 15,
559
+ "minimum_relative_root_polygon_slack": "0.982646291194200954546743206538",
560
+ "precision_decimal_digits": 90,
561
+ "arithmetic": "high-precision numerical; not an exact certificate"
562
+ },
563
+ {
564
+ "trial": 2,
565
+ "input_parameter": "1",
566
+ "output_parameter": "1.0",
567
+ "stage": "first",
568
+ "quartet_count": 35,
569
+ "minimum_relative_root_polygon_slack": "0.80229308308595869641546171279",
570
+ "precision_decimal_digits": 90,
571
+ "arithmetic": "high-precision numerical; not an exact certificate"
572
+ },
573
+ {
574
+ "trial": 2,
575
+ "input_parameter": "1",
576
+ "output_parameter": "1.0",
577
+ "stage": "composed",
578
+ "quartet_count": 15,
579
+ "minimum_relative_root_polygon_slack": "0.983155031121008271536844932803",
580
+ "precision_decimal_digits": 90,
581
+ "arithmetic": "high-precision numerical; not an exact certificate"
582
+ },
583
+ {
584
+ "trial": 2,
585
+ "input_parameter": "1.5",
586
+ "output_parameter": "1.5",
587
+ "stage": "first",
588
+ "quartet_count": 35,
589
+ "minimum_relative_root_polygon_slack": "0.829079462911511002076762574541",
590
+ "precision_decimal_digits": 90,
591
+ "arithmetic": "high-precision numerical; not an exact certificate"
592
+ },
593
+ {
594
+ "trial": 2,
595
+ "input_parameter": "1.5",
596
+ "output_parameter": "1.5",
597
+ "stage": "composed",
598
+ "quartet_count": 15,
599
+ "minimum_relative_root_polygon_slack": "0.985879408539973167498118261149",
600
+ "precision_decimal_digits": 90,
601
+ "arithmetic": "high-precision numerical; not an exact certificate"
602
+ },
603
+ {
604
+ "trial": 2,
605
+ "input_parameter": "2",
606
+ "output_parameter": "2.0",
607
+ "stage": "first",
608
+ "quartet_count": 35,
609
+ "minimum_relative_root_polygon_slack": "0.857532706842104873176769623034",
610
+ "precision_decimal_digits": 90,
611
+ "arithmetic": "high-precision numerical; not an exact certificate"
612
+ },
613
+ {
614
+ "trial": 2,
615
+ "input_parameter": "2",
616
+ "output_parameter": "2.0",
617
+ "stage": "composed",
618
+ "quartet_count": 15,
619
+ "minimum_relative_root_polygon_slack": "0.987119049058210851815854121009",
620
+ "precision_decimal_digits": 90,
621
+ "arithmetic": "high-precision numerical; not an exact certificate"
622
+ },
623
+ {
624
+ "trial": 2,
625
+ "input_parameter": "3",
626
+ "output_parameter": "3.0",
627
+ "stage": "first",
628
+ "quartet_count": 35,
629
+ "minimum_relative_root_polygon_slack": "0.922013238175791193494788747754",
630
+ "precision_decimal_digits": 90,
631
+ "arithmetic": "high-precision numerical; not an exact certificate"
632
+ },
633
+ {
634
+ "trial": 2,
635
+ "input_parameter": "3",
636
+ "output_parameter": "3.0",
637
+ "stage": "composed",
638
+ "quartet_count": 15,
639
+ "minimum_relative_root_polygon_slack": "0.988223130113206074755250382207",
640
+ "precision_decimal_digits": 90,
641
+ "arithmetic": "high-precision numerical; not an exact certificate"
642
+ },
643
+ {
644
+ "trial": 2,
645
+ "input_parameter": "4",
646
+ "output_parameter": "4.0",
647
+ "stage": "first",
648
+ "quartet_count": 35,
649
+ "minimum_relative_root_polygon_slack": "0.945127484511558321332776163735",
650
+ "precision_decimal_digits": 90,
651
+ "arithmetic": "high-precision numerical; not an exact certificate"
652
+ },
653
+ {
654
+ "trial": 2,
655
+ "input_parameter": "4",
656
+ "output_parameter": "4.0",
657
+ "stage": "composed",
658
+ "quartet_count": 15,
659
+ "minimum_relative_root_polygon_slack": "0.982988857205165443579476809932",
660
+ "precision_decimal_digits": 90,
661
+ "arithmetic": "high-precision numerical; not an exact certificate"
662
+ },
663
+ {
664
+ "trial": 2,
665
+ "input_parameter": "8",
666
+ "output_parameter": "8.0",
667
+ "stage": "first",
668
+ "quartet_count": 35,
669
+ "minimum_relative_root_polygon_slack": "0.862031986947695696959602075377",
670
+ "precision_decimal_digits": 90,
671
+ "arithmetic": "high-precision numerical; not an exact certificate"
672
+ },
673
+ {
674
+ "trial": 2,
675
+ "input_parameter": "8",
676
+ "output_parameter": "8.0",
677
+ "stage": "composed",
678
+ "quartet_count": 15,
679
+ "minimum_relative_root_polygon_slack": "0.969702364852670997220848107159",
680
+ "precision_decimal_digits": 90,
681
+ "arithmetic": "high-precision numerical; not an exact certificate"
682
+ },
683
+ {
684
+ "trial": 3,
685
+ "input_parameter": "0.25",
686
+ "output_parameter": "1.0",
687
+ "stage": "first",
688
+ "quartet_count": 35,
689
+ "minimum_relative_root_polygon_slack": "0.320626410417116426806594101976",
690
+ "precision_decimal_digits": 90,
691
+ "arithmetic": "high-precision numerical; not an exact certificate"
692
+ },
693
+ {
694
+ "trial": 3,
695
+ "input_parameter": "0.25",
696
+ "output_parameter": "1.0",
697
+ "stage": "composed",
698
+ "quartet_count": 15,
699
+ "minimum_relative_root_polygon_slack": "0.825758955745530130861640976553",
700
+ "precision_decimal_digits": 90,
701
+ "arithmetic": "high-precision numerical; not an exact certificate"
702
+ },
703
+ {
704
+ "trial": 3,
705
+ "input_parameter": "0.5",
706
+ "output_parameter": "1.0",
707
+ "stage": "first",
708
+ "quartet_count": 35,
709
+ "minimum_relative_root_polygon_slack": "0.340885824766589029142881316158",
710
+ "precision_decimal_digits": 90,
711
+ "arithmetic": "high-precision numerical; not an exact certificate"
712
+ },
713
+ {
714
+ "trial": 3,
715
+ "input_parameter": "0.5",
716
+ "output_parameter": "1.0",
717
+ "stage": "composed",
718
+ "quartet_count": 15,
719
+ "minimum_relative_root_polygon_slack": "0.850871880531386927962790922114",
720
+ "precision_decimal_digits": 90,
721
+ "arithmetic": "high-precision numerical; not an exact certificate"
722
+ },
723
+ {
724
+ "trial": 3,
725
+ "input_parameter": "1",
726
+ "output_parameter": "1.0",
727
+ "stage": "first",
728
+ "quartet_count": 35,
729
+ "minimum_relative_root_polygon_slack": "0.245934506571619514368456226331",
730
+ "precision_decimal_digits": 90,
731
+ "arithmetic": "high-precision numerical; not an exact certificate"
732
+ },
733
+ {
734
+ "trial": 3,
735
+ "input_parameter": "1",
736
+ "output_parameter": "1.0",
737
+ "stage": "composed",
738
+ "quartet_count": 15,
739
+ "minimum_relative_root_polygon_slack": "0.870931002508270959314190773004",
740
+ "precision_decimal_digits": 90,
741
+ "arithmetic": "high-precision numerical; not an exact certificate"
742
+ },
743
+ {
744
+ "trial": 3,
745
+ "input_parameter": "1.5",
746
+ "output_parameter": "1.5",
747
+ "stage": "first",
748
+ "quartet_count": 35,
749
+ "minimum_relative_root_polygon_slack": "0.298573563720075948471784172271",
750
+ "precision_decimal_digits": 90,
751
+ "arithmetic": "high-precision numerical; not an exact certificate"
752
+ },
753
+ {
754
+ "trial": 3,
755
+ "input_parameter": "1.5",
756
+ "output_parameter": "1.5",
757
+ "stage": "composed",
758
+ "quartet_count": 15,
759
+ "minimum_relative_root_polygon_slack": "0.895162240800689647027980176752",
760
+ "precision_decimal_digits": 90,
761
+ "arithmetic": "high-precision numerical; not an exact certificate"
762
+ },
763
+ {
764
+ "trial": 3,
765
+ "input_parameter": "2",
766
+ "output_parameter": "2.0",
767
+ "stage": "first",
768
+ "quartet_count": 35,
769
+ "minimum_relative_root_polygon_slack": "0.318491796619548100521289322136",
770
+ "precision_decimal_digits": 90,
771
+ "arithmetic": "high-precision numerical; not an exact certificate"
772
+ },
773
+ {
774
+ "trial": 3,
775
+ "input_parameter": "2",
776
+ "output_parameter": "2.0",
777
+ "stage": "composed",
778
+ "quartet_count": 15,
779
+ "minimum_relative_root_polygon_slack": "0.890153494263936990453484519355",
780
+ "precision_decimal_digits": 90,
781
+ "arithmetic": "high-precision numerical; not an exact certificate"
782
+ },
783
+ {
784
+ "trial": 3,
785
+ "input_parameter": "3",
786
+ "output_parameter": "3.0",
787
+ "stage": "first",
788
+ "quartet_count": 35,
789
+ "minimum_relative_root_polygon_slack": "0.325760601108534500670184129704",
790
+ "precision_decimal_digits": 90,
791
+ "arithmetic": "high-precision numerical; not an exact certificate"
792
+ },
793
+ {
794
+ "trial": 3,
795
+ "input_parameter": "3",
796
+ "output_parameter": "3.0",
797
+ "stage": "composed",
798
+ "quartet_count": 15,
799
+ "minimum_relative_root_polygon_slack": "0.874833828177910775517509006308",
800
+ "precision_decimal_digits": 90,
801
+ "arithmetic": "high-precision numerical; not an exact certificate"
802
+ },
803
+ {
804
+ "trial": 3,
805
+ "input_parameter": "4",
806
+ "output_parameter": "4.0",
807
+ "stage": "first",
808
+ "quartet_count": 35,
809
+ "minimum_relative_root_polygon_slack": "0.322199980938715101126002303273",
810
+ "precision_decimal_digits": 90,
811
+ "arithmetic": "high-precision numerical; not an exact certificate"
812
+ },
813
+ {
814
+ "trial": 3,
815
+ "input_parameter": "4",
816
+ "output_parameter": "4.0",
817
+ "stage": "composed",
818
+ "quartet_count": 15,
819
+ "minimum_relative_root_polygon_slack": "0.875322307351029243286837702663",
820
+ "precision_decimal_digits": 90,
821
+ "arithmetic": "high-precision numerical; not an exact certificate"
822
+ },
823
+ {
824
+ "trial": 3,
825
+ "input_parameter": "8",
826
+ "output_parameter": "8.0",
827
+ "stage": "first",
828
+ "quartet_count": 35,
829
+ "minimum_relative_root_polygon_slack": "0.30935010057925781063032826994",
830
+ "precision_decimal_digits": 90,
831
+ "arithmetic": "high-precision numerical; not an exact certificate"
832
+ },
833
+ {
834
+ "trial": 3,
835
+ "input_parameter": "8",
836
+ "output_parameter": "8.0",
837
+ "stage": "composed",
838
+ "quartet_count": 15,
839
+ "minimum_relative_root_polygon_slack": "0.916893236494143532115361908703",
840
+ "precision_decimal_digits": 90,
841
+ "arithmetic": "high-precision numerical; not an exact certificate"
842
+ },
843
+ {
844
+ "trial": 4,
845
+ "input_parameter": "0.25",
846
+ "output_parameter": "1.0",
847
+ "stage": "first",
848
+ "quartet_count": 35,
849
+ "minimum_relative_root_polygon_slack": "0.346138275324873787455715671778",
850
+ "precision_decimal_digits": 90,
851
+ "arithmetic": "high-precision numerical; not an exact certificate"
852
+ },
853
+ {
854
+ "trial": 4,
855
+ "input_parameter": "0.25",
856
+ "output_parameter": "1.0",
857
+ "stage": "composed",
858
+ "quartet_count": 15,
859
+ "minimum_relative_root_polygon_slack": "0.803972065682679632548252335526",
860
+ "precision_decimal_digits": 90,
861
+ "arithmetic": "high-precision numerical; not an exact certificate"
862
+ },
863
+ {
864
+ "trial": 4,
865
+ "input_parameter": "0.5",
866
+ "output_parameter": "1.0",
867
+ "stage": "first",
868
+ "quartet_count": 35,
869
+ "minimum_relative_root_polygon_slack": "0.308873978049598685736871434715",
870
+ "precision_decimal_digits": 90,
871
+ "arithmetic": "high-precision numerical; not an exact certificate"
872
+ },
873
+ {
874
+ "trial": 4,
875
+ "input_parameter": "0.5",
876
+ "output_parameter": "1.0",
877
+ "stage": "composed",
878
+ "quartet_count": 15,
879
+ "minimum_relative_root_polygon_slack": "0.791945737959673059985392034439",
880
+ "precision_decimal_digits": 90,
881
+ "arithmetic": "high-precision numerical; not an exact certificate"
882
+ },
883
+ {
884
+ "trial": 4,
885
+ "input_parameter": "1",
886
+ "output_parameter": "1.0",
887
+ "stage": "first",
888
+ "quartet_count": 35,
889
+ "minimum_relative_root_polygon_slack": "0.238754325259515570934256055363",
890
+ "precision_decimal_digits": 90,
891
+ "arithmetic": "high-precision numerical; not an exact certificate"
892
+ },
893
+ {
894
+ "trial": 4,
895
+ "input_parameter": "1",
896
+ "output_parameter": "1.0",
897
+ "stage": "composed",
898
+ "quartet_count": 15,
899
+ "minimum_relative_root_polygon_slack": "0.766282180606113766940046489189",
900
+ "precision_decimal_digits": 90,
901
+ "arithmetic": "high-precision numerical; not an exact certificate"
902
+ },
903
+ {
904
+ "trial": 4,
905
+ "input_parameter": "1.5",
906
+ "output_parameter": "1.5",
907
+ "stage": "first",
908
+ "quartet_count": 35,
909
+ "minimum_relative_root_polygon_slack": "0.316373016752334629317669515086",
910
+ "precision_decimal_digits": 90,
911
+ "arithmetic": "high-precision numerical; not an exact certificate"
912
+ },
913
+ {
914
+ "trial": 4,
915
+ "input_parameter": "1.5",
916
+ "output_parameter": "1.5",
917
+ "stage": "composed",
918
+ "quartet_count": 15,
919
+ "minimum_relative_root_polygon_slack": "0.817854295188449050117772710437",
920
+ "precision_decimal_digits": 90,
921
+ "arithmetic": "high-precision numerical; not an exact certificate"
922
+ },
923
+ {
924
+ "trial": 4,
925
+ "input_parameter": "2",
926
+ "output_parameter": "2.0",
927
+ "stage": "first",
928
+ "quartet_count": 35,
929
+ "minimum_relative_root_polygon_slack": "0.35874534819572365559824905932",
930
+ "precision_decimal_digits": 90,
931
+ "arithmetic": "high-precision numerical; not an exact certificate"
932
+ },
933
+ {
934
+ "trial": 4,
935
+ "input_parameter": "2",
936
+ "output_parameter": "2.0",
937
+ "stage": "composed",
938
+ "quartet_count": 15,
939
+ "minimum_relative_root_polygon_slack": "0.844714413816744701675216072577",
940
+ "precision_decimal_digits": 90,
941
+ "arithmetic": "high-precision numerical; not an exact certificate"
942
+ },
943
+ {
944
+ "trial": 4,
945
+ "input_parameter": "3",
946
+ "output_parameter": "3.0",
947
+ "stage": "first",
948
+ "quartet_count": 35,
949
+ "minimum_relative_root_polygon_slack": "0.399024828389551567936665335974",
950
+ "precision_decimal_digits": 90,
951
+ "arithmetic": "high-precision numerical; not an exact certificate"
952
+ },
953
+ {
954
+ "trial": 4,
955
+ "input_parameter": "3",
956
+ "output_parameter": "3.0",
957
+ "stage": "composed",
958
+ "quartet_count": 15,
959
+ "minimum_relative_root_polygon_slack": "0.874323350312073503538470971388",
960
+ "precision_decimal_digits": 90,
961
+ "arithmetic": "high-precision numerical; not an exact certificate"
962
+ },
963
+ {
964
+ "trial": 4,
965
+ "input_parameter": "4",
966
+ "output_parameter": "4.0",
967
+ "stage": "first",
968
+ "quartet_count": 35,
969
+ "minimum_relative_root_polygon_slack": "0.417795405129247160589097704705",
970
+ "precision_decimal_digits": 90,
971
+ "arithmetic": "high-precision numerical; not an exact certificate"
972
+ },
973
+ {
974
+ "trial": 4,
975
+ "input_parameter": "4",
976
+ "output_parameter": "4.0",
977
+ "stage": "composed",
978
+ "quartet_count": 15,
979
+ "minimum_relative_root_polygon_slack": "0.893253199586415267246658874457",
980
+ "precision_decimal_digits": 90,
981
+ "arithmetic": "high-precision numerical; not an exact certificate"
982
+ },
983
+ {
984
+ "trial": 4,
985
+ "input_parameter": "8",
986
+ "output_parameter": "8.0",
987
+ "stage": "first",
988
+ "quartet_count": 35,
989
+ "minimum_relative_root_polygon_slack": "0.452256030562380140919751740104",
990
+ "precision_decimal_digits": 90,
991
+ "arithmetic": "high-precision numerical; not an exact certificate"
992
+ },
993
+ {
994
+ "trial": 4,
995
+ "input_parameter": "8",
996
+ "output_parameter": "8.0",
997
+ "stage": "composed",
998
+ "quartet_count": 15,
999
+ "minimum_relative_root_polygon_slack": "0.937754695285110560806687917746",
1000
+ "precision_decimal_digits": 90,
1001
+ "arithmetic": "high-precision numerical; not an exact certificate"
1002
+ },
1003
+ {
1004
+ "trial": 5,
1005
+ "input_parameter": "0.25",
1006
+ "output_parameter": "1.0",
1007
+ "stage": "first",
1008
+ "quartet_count": 35,
1009
+ "minimum_relative_root_polygon_slack": "0.453293055083452278827648336178",
1010
+ "precision_decimal_digits": 90,
1011
+ "arithmetic": "high-precision numerical; not an exact certificate"
1012
+ },
1013
+ {
1014
+ "trial": 5,
1015
+ "input_parameter": "0.25",
1016
+ "output_parameter": "1.0",
1017
+ "stage": "composed",
1018
+ "quartet_count": 15,
1019
+ "minimum_relative_root_polygon_slack": "0.935051317804252240723743258797",
1020
+ "precision_decimal_digits": 90,
1021
+ "arithmetic": "high-precision numerical; not an exact certificate"
1022
+ },
1023
+ {
1024
+ "trial": 5,
1025
+ "input_parameter": "0.5",
1026
+ "output_parameter": "1.0",
1027
+ "stage": "first",
1028
+ "quartet_count": 35,
1029
+ "minimum_relative_root_polygon_slack": "0.400936630742398342461213970274",
1030
+ "precision_decimal_digits": 90,
1031
+ "arithmetic": "high-precision numerical; not an exact certificate"
1032
+ },
1033
+ {
1034
+ "trial": 5,
1035
+ "input_parameter": "0.5",
1036
+ "output_parameter": "1.0",
1037
+ "stage": "composed",
1038
+ "quartet_count": 15,
1039
+ "minimum_relative_root_polygon_slack": "0.910228122269169161978806772281",
1040
+ "precision_decimal_digits": 90,
1041
+ "arithmetic": "high-precision numerical; not an exact certificate"
1042
+ },
1043
+ {
1044
+ "trial": 5,
1045
+ "input_parameter": "1",
1046
+ "output_parameter": "1.0",
1047
+ "stage": "first",
1048
+ "quartet_count": 35,
1049
+ "minimum_relative_root_polygon_slack": "0.25974025974025974025974025974",
1050
+ "precision_decimal_digits": 90,
1051
+ "arithmetic": "high-precision numerical; not an exact certificate"
1052
+ },
1053
+ {
1054
+ "trial": 5,
1055
+ "input_parameter": "1",
1056
+ "output_parameter": "1.0",
1057
+ "stage": "composed",
1058
+ "quartet_count": 15,
1059
+ "minimum_relative_root_polygon_slack": "0.856074308558855706392127823852",
1060
+ "precision_decimal_digits": 90,
1061
+ "arithmetic": "high-precision numerical; not an exact certificate"
1062
+ },
1063
+ {
1064
+ "trial": 5,
1065
+ "input_parameter": "1.5",
1066
+ "output_parameter": "1.5",
1067
+ "stage": "first",
1068
+ "quartet_count": 35,
1069
+ "minimum_relative_root_polygon_slack": "0.255359407473861844949141529231",
1070
+ "precision_decimal_digits": 90,
1071
+ "arithmetic": "high-precision numerical; not an exact certificate"
1072
+ },
1073
+ {
1074
+ "trial": 5,
1075
+ "input_parameter": "1.5",
1076
+ "output_parameter": "1.5",
1077
+ "stage": "composed",
1078
+ "quartet_count": 15,
1079
+ "minimum_relative_root_polygon_slack": "0.866593251334669171575235636673",
1080
+ "precision_decimal_digits": 90,
1081
+ "arithmetic": "high-precision numerical; not an exact certificate"
1082
+ },
1083
+ {
1084
+ "trial": 5,
1085
+ "input_parameter": "2",
1086
+ "output_parameter": "2.0",
1087
+ "stage": "first",
1088
+ "quartet_count": 35,
1089
+ "minimum_relative_root_polygon_slack": "0.229064118469851952433705730298",
1090
+ "precision_decimal_digits": 90,
1091
+ "arithmetic": "high-precision numerical; not an exact certificate"
1092
+ },
1093
+ {
1094
+ "trial": 5,
1095
+ "input_parameter": "2",
1096
+ "output_parameter": "2.0",
1097
+ "stage": "composed",
1098
+ "quartet_count": 15,
1099
+ "minimum_relative_root_polygon_slack": "0.876389629528637276828325884904",
1100
+ "precision_decimal_digits": 90,
1101
+ "arithmetic": "high-precision numerical; not an exact certificate"
1102
+ },
1103
+ {
1104
+ "trial": 5,
1105
+ "input_parameter": "3",
1106
+ "output_parameter": "3.0",
1107
+ "stage": "first",
1108
+ "quartet_count": 35,
1109
+ "minimum_relative_root_polygon_slack": "0.188610984531242434490017608432",
1110
+ "precision_decimal_digits": 90,
1111
+ "arithmetic": "high-precision numerical; not an exact certificate"
1112
+ },
1113
+ {
1114
+ "trial": 5,
1115
+ "input_parameter": "3",
1116
+ "output_parameter": "3.0",
1117
+ "stage": "composed",
1118
+ "quartet_count": 15,
1119
+ "minimum_relative_root_polygon_slack": "0.894818124619657687309663344062",
1120
+ "precision_decimal_digits": 90,
1121
+ "arithmetic": "high-precision numerical; not an exact certificate"
1122
+ },
1123
+ {
1124
+ "trial": 5,
1125
+ "input_parameter": "4",
1126
+ "output_parameter": "4.0",
1127
+ "stage": "first",
1128
+ "quartet_count": 35,
1129
+ "minimum_relative_root_polygon_slack": "0.168850202288152072224880297868",
1130
+ "precision_decimal_digits": 90,
1131
+ "arithmetic": "high-precision numerical; not an exact certificate"
1132
+ },
1133
+ {
1134
+ "trial": 5,
1135
+ "input_parameter": "4",
1136
+ "output_parameter": "4.0",
1137
+ "stage": "composed",
1138
+ "quartet_count": 15,
1139
+ "minimum_relative_root_polygon_slack": "0.909247112717539034739877171651",
1140
+ "precision_decimal_digits": 90,
1141
+ "arithmetic": "high-precision numerical; not an exact certificate"
1142
+ },
1143
+ {
1144
+ "trial": 5,
1145
+ "input_parameter": "8",
1146
+ "output_parameter": "8.0",
1147
+ "stage": "first",
1148
+ "quartet_count": 35,
1149
+ "minimum_relative_root_polygon_slack": "0.148836464688231677693148218232",
1150
+ "precision_decimal_digits": 90,
1151
+ "arithmetic": "high-precision numerical; not an exact certificate"
1152
+ },
1153
+ {
1154
+ "trial": 5,
1155
+ "input_parameter": "8",
1156
+ "output_parameter": "8.0",
1157
+ "stage": "composed",
1158
+ "quartet_count": 15,
1159
+ "minimum_relative_root_polygon_slack": "0.943140788016417624712642094739",
1160
+ "precision_decimal_digits": 90,
1161
+ "arithmetic": "high-precision numerical; not an exact certificate"
1162
+ },
1163
+ {
1164
+ "trial": 6,
1165
+ "input_parameter": "0.25",
1166
+ "output_parameter": "1.0",
1167
+ "stage": "first",
1168
+ "quartet_count": 35,
1169
+ "minimum_relative_root_polygon_slack": "0.481983089393763676952312587347",
1170
+ "precision_decimal_digits": 90,
1171
+ "arithmetic": "high-precision numerical; not an exact certificate"
1172
+ },
1173
+ {
1174
+ "trial": 6,
1175
+ "input_parameter": "0.25",
1176
+ "output_parameter": "1.0",
1177
+ "stage": "composed",
1178
+ "quartet_count": 15,
1179
+ "minimum_relative_root_polygon_slack": "0.845937080077566244808735675866",
1180
+ "precision_decimal_digits": 90,
1181
+ "arithmetic": "high-precision numerical; not an exact certificate"
1182
+ },
1183
+ {
1184
+ "trial": 6,
1185
+ "input_parameter": "0.5",
1186
+ "output_parameter": "1.0",
1187
+ "stage": "first",
1188
+ "quartet_count": 35,
1189
+ "minimum_relative_root_polygon_slack": "0.459287512860692938193268712282",
1190
+ "precision_decimal_digits": 90,
1191
+ "arithmetic": "high-precision numerical; not an exact certificate"
1192
+ },
1193
+ {
1194
+ "trial": 6,
1195
+ "input_parameter": "0.5",
1196
+ "output_parameter": "1.0",
1197
+ "stage": "composed",
1198
+ "quartet_count": 15,
1199
+ "minimum_relative_root_polygon_slack": "0.842580683652624147268491178231",
1200
+ "precision_decimal_digits": 90,
1201
+ "arithmetic": "high-precision numerical; not an exact certificate"
1202
+ },
1203
+ {
1204
+ "trial": 6,
1205
+ "input_parameter": "1",
1206
+ "output_parameter": "1.0",
1207
+ "stage": "first",
1208
+ "quartet_count": 35,
1209
+ "minimum_relative_root_polygon_slack": "0.379828900368313085130181641662",
1210
+ "precision_decimal_digits": 90,
1211
+ "arithmetic": "high-precision numerical; not an exact certificate"
1212
+ },
1213
+ {
1214
+ "trial": 6,
1215
+ "input_parameter": "1",
1216
+ "output_parameter": "1.0",
1217
+ "stage": "composed",
1218
+ "quartet_count": 15,
1219
+ "minimum_relative_root_polygon_slack": "0.847626211862856790842185421588",
1220
+ "precision_decimal_digits": 90,
1221
+ "arithmetic": "high-precision numerical; not an exact certificate"
1222
+ },
1223
+ {
1224
+ "trial": 6,
1225
+ "input_parameter": "1.5",
1226
+ "output_parameter": "1.5",
1227
+ "stage": "first",
1228
+ "quartet_count": 35,
1229
+ "minimum_relative_root_polygon_slack": "0.469784364277617108513536713561",
1230
+ "precision_decimal_digits": 90,
1231
+ "arithmetic": "high-precision numerical; not an exact certificate"
1232
+ },
1233
+ {
1234
+ "trial": 6,
1235
+ "input_parameter": "1.5",
1236
+ "output_parameter": "1.5",
1237
+ "stage": "composed",
1238
+ "quartet_count": 15,
1239
+ "minimum_relative_root_polygon_slack": "0.907743432085327332262493348292",
1240
+ "precision_decimal_digits": 90,
1241
+ "arithmetic": "high-precision numerical; not an exact certificate"
1242
+ },
1243
+ {
1244
+ "trial": 6,
1245
+ "input_parameter": "2",
1246
+ "output_parameter": "2.0",
1247
+ "stage": "first",
1248
+ "quartet_count": 35,
1249
+ "minimum_relative_root_polygon_slack": "0.525611848215037920843458050735",
1250
+ "precision_decimal_digits": 90,
1251
+ "arithmetic": "high-precision numerical; not an exact certificate"
1252
+ },
1253
+ {
1254
+ "trial": 6,
1255
+ "input_parameter": "2",
1256
+ "output_parameter": "2.0",
1257
+ "stage": "composed",
1258
+ "quartet_count": 15,
1259
+ "minimum_relative_root_polygon_slack": "0.942653544024915451500930573805",
1260
+ "precision_decimal_digits": 90,
1261
+ "arithmetic": "high-precision numerical; not an exact certificate"
1262
+ },
1263
+ {
1264
+ "trial": 6,
1265
+ "input_parameter": "3",
1266
+ "output_parameter": "3.0",
1267
+ "stage": "first",
1268
+ "quartet_count": 35,
1269
+ "minimum_relative_root_polygon_slack": "0.567091013363177952346896527716",
1270
+ "precision_decimal_digits": 90,
1271
+ "arithmetic": "high-precision numerical; not an exact certificate"
1272
+ },
1273
+ {
1274
+ "trial": 6,
1275
+ "input_parameter": "3",
1276
+ "output_parameter": "3.0",
1277
+ "stage": "composed",
1278
+ "quartet_count": 15,
1279
+ "minimum_relative_root_polygon_slack": "0.978406096444485635419805873252",
1280
+ "precision_decimal_digits": 90,
1281
+ "arithmetic": "high-precision numerical; not an exact certificate"
1282
+ },
1283
+ {
1284
+ "trial": 6,
1285
+ "input_parameter": "4",
1286
+ "output_parameter": "4.0",
1287
+ "stage": "first",
1288
+ "quartet_count": 35,
1289
+ "minimum_relative_root_polygon_slack": "0.520439401887972622135559612556",
1290
+ "precision_decimal_digits": 90,
1291
+ "arithmetic": "high-precision numerical; not an exact certificate"
1292
+ },
1293
+ {
1294
+ "trial": 6,
1295
+ "input_parameter": "4",
1296
+ "output_parameter": "4.0",
1297
+ "stage": "composed",
1298
+ "quartet_count": 15,
1299
+ "minimum_relative_root_polygon_slack": "0.992425554999097708733820917068",
1300
+ "precision_decimal_digits": 90,
1301
+ "arithmetic": "high-precision numerical; not an exact certificate"
1302
+ },
1303
+ {
1304
+ "trial": 6,
1305
+ "input_parameter": "8",
1306
+ "output_parameter": "8.0",
1307
+ "stage": "first",
1308
+ "quartet_count": 35,
1309
+ "minimum_relative_root_polygon_slack": "0.408561707602828439467928227355",
1310
+ "precision_decimal_digits": 90,
1311
+ "arithmetic": "high-precision numerical; not an exact certificate"
1312
+ },
1313
+ {
1314
+ "trial": 6,
1315
+ "input_parameter": "8",
1316
+ "output_parameter": "8.0",
1317
+ "stage": "composed",
1318
+ "quartet_count": 15,
1319
+ "minimum_relative_root_polygon_slack": "0.99990033363577889147246150375",
1320
+ "precision_decimal_digits": 90,
1321
+ "arithmetic": "high-precision numerical; not an exact certificate"
1322
+ },
1323
+ {
1324
+ "trial": 7,
1325
+ "input_parameter": "0.25",
1326
+ "output_parameter": "1.0",
1327
+ "stage": "first",
1328
+ "quartet_count": 35,
1329
+ "minimum_relative_root_polygon_slack": "0.0",
1330
+ "precision_decimal_digits": 90,
1331
+ "arithmetic": "high-precision numerical; not an exact certificate"
1332
+ },
1333
+ {
1334
+ "trial": 7,
1335
+ "input_parameter": "0.25",
1336
+ "output_parameter": "1.0",
1337
+ "stage": "composed",
1338
+ "quartet_count": 15,
1339
+ "minimum_relative_root_polygon_slack": "0.767941308613197062184914532201",
1340
+ "precision_decimal_digits": 90,
1341
+ "arithmetic": "high-precision numerical; not an exact certificate"
1342
+ },
1343
+ {
1344
+ "trial": 7,
1345
+ "input_parameter": "0.5",
1346
+ "output_parameter": "1.0",
1347
+ "stage": "first",
1348
+ "quartet_count": 35,
1349
+ "minimum_relative_root_polygon_slack": "0.0",
1350
+ "precision_decimal_digits": 90,
1351
+ "arithmetic": "high-precision numerical; not an exact certificate"
1352
+ },
1353
+ {
1354
+ "trial": 7,
1355
+ "input_parameter": "0.5",
1356
+ "output_parameter": "1.0",
1357
+ "stage": "composed",
1358
+ "quartet_count": 15,
1359
+ "minimum_relative_root_polygon_slack": "0.738048648945793349932749343375",
1360
+ "precision_decimal_digits": 90,
1361
+ "arithmetic": "high-precision numerical; not an exact certificate"
1362
+ },
1363
+ {
1364
+ "trial": 7,
1365
+ "input_parameter": "1",
1366
+ "output_parameter": "1.0",
1367
+ "stage": "first",
1368
+ "quartet_count": 35,
1369
+ "minimum_relative_root_polygon_slack": "0.0",
1370
+ "precision_decimal_digits": 90,
1371
+ "arithmetic": "high-precision numerical; not an exact certificate"
1372
+ },
1373
+ {
1374
+ "trial": 7,
1375
+ "input_parameter": "1",
1376
+ "output_parameter": "1.0",
1377
+ "stage": "composed",
1378
+ "quartet_count": 15,
1379
+ "minimum_relative_root_polygon_slack": "0.672747032486463973344439816743",
1380
+ "precision_decimal_digits": 90,
1381
+ "arithmetic": "high-precision numerical; not an exact certificate"
1382
+ },
1383
+ {
1384
+ "trial": 7,
1385
+ "input_parameter": "1.5",
1386
+ "output_parameter": "1.5",
1387
+ "stage": "first",
1388
+ "quartet_count": 35,
1389
+ "minimum_relative_root_polygon_slack": "0.0",
1390
+ "precision_decimal_digits": 90,
1391
+ "arithmetic": "high-precision numerical; not an exact certificate"
1392
+ },
1393
+ {
1394
+ "trial": 7,
1395
+ "input_parameter": "1.5",
1396
+ "output_parameter": "1.5",
1397
+ "stage": "composed",
1398
+ "quartet_count": 15,
1399
+ "minimum_relative_root_polygon_slack": "0.713635556177332718601458492811",
1400
+ "precision_decimal_digits": 90,
1401
+ "arithmetic": "high-precision numerical; not an exact certificate"
1402
+ },
1403
+ {
1404
+ "trial": 7,
1405
+ "input_parameter": "2",
1406
+ "output_parameter": "2.0",
1407
+ "stage": "first",
1408
+ "quartet_count": 35,
1409
+ "minimum_relative_root_polygon_slack": "0.0",
1410
+ "precision_decimal_digits": 90,
1411
+ "arithmetic": "high-precision numerical; not an exact certificate"
1412
+ },
1413
+ {
1414
+ "trial": 7,
1415
+ "input_parameter": "2",
1416
+ "output_parameter": "2.0",
1417
+ "stage": "composed",
1418
+ "quartet_count": 15,
1419
+ "minimum_relative_root_polygon_slack": "0.729814405887515042832482791853",
1420
+ "precision_decimal_digits": 90,
1421
+ "arithmetic": "high-precision numerical; not an exact certificate"
1422
+ },
1423
+ {
1424
+ "trial": 7,
1425
+ "input_parameter": "3",
1426
+ "output_parameter": "3.0",
1427
+ "stage": "first",
1428
+ "quartet_count": 35,
1429
+ "minimum_relative_root_polygon_slack": "0.0",
1430
+ "precision_decimal_digits": 90,
1431
+ "arithmetic": "high-precision numerical; not an exact certificate"
1432
+ },
1433
+ {
1434
+ "trial": 7,
1435
+ "input_parameter": "3",
1436
+ "output_parameter": "3.0",
1437
+ "stage": "composed",
1438
+ "quartet_count": 15,
1439
+ "minimum_relative_root_polygon_slack": "0.741605202339751946826999769562",
1440
+ "precision_decimal_digits": 90,
1441
+ "arithmetic": "high-precision numerical; not an exact certificate"
1442
+ },
1443
+ {
1444
+ "trial": 7,
1445
+ "input_parameter": "4",
1446
+ "output_parameter": "4.0",
1447
+ "stage": "first",
1448
+ "quartet_count": 35,
1449
+ "minimum_relative_root_polygon_slack": "0.0",
1450
+ "precision_decimal_digits": 90,
1451
+ "arithmetic": "high-precision numerical; not an exact certificate"
1452
+ },
1453
+ {
1454
+ "trial": 7,
1455
+ "input_parameter": "4",
1456
+ "output_parameter": "4.0",
1457
+ "stage": "composed",
1458
+ "quartet_count": 15,
1459
+ "minimum_relative_root_polygon_slack": "0.751757817202688794002463815939",
1460
+ "precision_decimal_digits": 90,
1461
+ "arithmetic": "high-precision numerical; not an exact certificate"
1462
+ },
1463
+ {
1464
+ "trial": 7,
1465
+ "input_parameter": "8",
1466
+ "output_parameter": "8.0",
1467
+ "stage": "first",
1468
+ "quartet_count": 35,
1469
+ "minimum_relative_root_polygon_slack": "0.0",
1470
+ "precision_decimal_digits": 90,
1471
+ "arithmetic": "high-precision numerical; not an exact certificate"
1472
+ },
1473
+ {
1474
+ "trial": 7,
1475
+ "input_parameter": "8",
1476
+ "output_parameter": "8.0",
1477
+ "stage": "composed",
1478
+ "quartet_count": 15,
1479
+ "minimum_relative_root_polygon_slack": "0.824667255263036872354476471284",
1480
+ "precision_decimal_digits": 90,
1481
+ "arithmetic": "high-precision numerical; not an exact certificate"
1482
+ },
1483
+ {
1484
+ "trial": 8,
1485
+ "input_parameter": "0.25",
1486
+ "output_parameter": "1.0",
1487
+ "stage": "first",
1488
+ "quartet_count": 35,
1489
+ "minimum_relative_root_polygon_slack": "0.343391947442442035183309134509",
1490
+ "precision_decimal_digits": 90,
1491
+ "arithmetic": "high-precision numerical; not an exact certificate"
1492
+ },
1493
+ {
1494
+ "trial": 8,
1495
+ "input_parameter": "0.25",
1496
+ "output_parameter": "1.0",
1497
+ "stage": "composed",
1498
+ "quartet_count": 15,
1499
+ "minimum_relative_root_polygon_slack": "0.938801514724179901598838575999",
1500
+ "precision_decimal_digits": 90,
1501
+ "arithmetic": "high-precision numerical; not an exact certificate"
1502
+ },
1503
+ {
1504
+ "trial": 8,
1505
+ "input_parameter": "0.5",
1506
+ "output_parameter": "1.0",
1507
+ "stage": "first",
1508
+ "quartet_count": 35,
1509
+ "minimum_relative_root_polygon_slack": "0.349288871067017744111760738532",
1510
+ "precision_decimal_digits": 90,
1511
+ "arithmetic": "high-precision numerical; not an exact certificate"
1512
+ },
1513
+ {
1514
+ "trial": 8,
1515
+ "input_parameter": "0.5",
1516
+ "output_parameter": "1.0",
1517
+ "stage": "composed",
1518
+ "quartet_count": 15,
1519
+ "minimum_relative_root_polygon_slack": "0.946779773780020264921620842243",
1520
+ "precision_decimal_digits": 90,
1521
+ "arithmetic": "high-precision numerical; not an exact certificate"
1522
+ },
1523
+ {
1524
+ "trial": 8,
1525
+ "input_parameter": "1",
1526
+ "output_parameter": "1.0",
1527
+ "stage": "first",
1528
+ "quartet_count": 35,
1529
+ "minimum_relative_root_polygon_slack": "0.356930398572278405710886377156",
1530
+ "precision_decimal_digits": 90,
1531
+ "arithmetic": "high-precision numerical; not an exact certificate"
1532
+ },
1533
+ {
1534
+ "trial": 8,
1535
+ "input_parameter": "1",
1536
+ "output_parameter": "1.0",
1537
+ "stage": "composed",
1538
+ "quartet_count": 15,
1539
+ "minimum_relative_root_polygon_slack": "0.964178524957248392530670540937",
1540
+ "precision_decimal_digits": 90,
1541
+ "arithmetic": "high-precision numerical; not an exact certificate"
1542
+ },
1543
+ {
1544
+ "trial": 8,
1545
+ "input_parameter": "1.5",
1546
+ "output_parameter": "1.5",
1547
+ "stage": "first",
1548
+ "quartet_count": 35,
1549
+ "minimum_relative_root_polygon_slack": "0.502970535030737045689900572079",
1550
+ "precision_decimal_digits": 90,
1551
+ "arithmetic": "high-precision numerical; not an exact certificate"
1552
+ },
1553
+ {
1554
+ "trial": 8,
1555
+ "input_parameter": "1.5",
1556
+ "output_parameter": "1.5",
1557
+ "stage": "composed",
1558
+ "quartet_count": 15,
1559
+ "minimum_relative_root_polygon_slack": "0.984752267672519359076504823032",
1560
+ "precision_decimal_digits": 90,
1561
+ "arithmetic": "high-precision numerical; not an exact certificate"
1562
+ },
1563
+ {
1564
+ "trial": 8,
1565
+ "input_parameter": "2",
1566
+ "output_parameter": "2.0",
1567
+ "stage": "first",
1568
+ "quartet_count": 35,
1569
+ "minimum_relative_root_polygon_slack": "0.514937397347175126307665190121",
1570
+ "precision_decimal_digits": 90,
1571
+ "arithmetic": "high-precision numerical; not an exact certificate"
1572
+ },
1573
+ {
1574
+ "trial": 8,
1575
+ "input_parameter": "2",
1576
+ "output_parameter": "2.0",
1577
+ "stage": "composed",
1578
+ "quartet_count": 15,
1579
+ "minimum_relative_root_polygon_slack": "0.987165476463838854582800560935",
1580
+ "precision_decimal_digits": 90,
1581
+ "arithmetic": "high-precision numerical; not an exact certificate"
1582
+ },
1583
+ {
1584
+ "trial": 8,
1585
+ "input_parameter": "3",
1586
+ "output_parameter": "3.0",
1587
+ "stage": "first",
1588
+ "quartet_count": 35,
1589
+ "minimum_relative_root_polygon_slack": "0.425670455285895749784997765928",
1590
+ "precision_decimal_digits": 90,
1591
+ "arithmetic": "high-precision numerical; not an exact certificate"
1592
+ },
1593
+ {
1594
+ "trial": 8,
1595
+ "input_parameter": "3",
1596
+ "output_parameter": "3.0",
1597
+ "stage": "composed",
1598
+ "quartet_count": 15,
1599
+ "minimum_relative_root_polygon_slack": "0.973166791476765092434130869843",
1600
+ "precision_decimal_digits": 90,
1601
+ "arithmetic": "high-precision numerical; not an exact certificate"
1602
+ },
1603
+ {
1604
+ "trial": 8,
1605
+ "input_parameter": "4",
1606
+ "output_parameter": "4.0",
1607
+ "stage": "first",
1608
+ "quartet_count": 35,
1609
+ "minimum_relative_root_polygon_slack": "0.357493766419380215638725138931",
1610
+ "precision_decimal_digits": 90,
1611
+ "arithmetic": "high-precision numerical; not an exact certificate"
1612
+ },
1613
+ {
1614
+ "trial": 8,
1615
+ "input_parameter": "4",
1616
+ "output_parameter": "4.0",
1617
+ "stage": "composed",
1618
+ "quartet_count": 15,
1619
+ "minimum_relative_root_polygon_slack": "0.965993536749990670371576850761",
1620
+ "precision_decimal_digits": 90,
1621
+ "arithmetic": "high-precision numerical; not an exact certificate"
1622
+ },
1623
+ {
1624
+ "trial": 8,
1625
+ "input_parameter": "8",
1626
+ "output_parameter": "8.0",
1627
+ "stage": "first",
1628
+ "quartet_count": 35,
1629
+ "minimum_relative_root_polygon_slack": "0.264727792545772755828485970896",
1630
+ "precision_decimal_digits": 90,
1631
+ "arithmetic": "high-precision numerical; not an exact certificate"
1632
+ },
1633
+ {
1634
+ "trial": 8,
1635
+ "input_parameter": "8",
1636
+ "output_parameter": "8.0",
1637
+ "stage": "composed",
1638
+ "quartet_count": 15,
1639
+ "minimum_relative_root_polygon_slack": "0.972905151089934240909491288981",
1640
+ "precision_decimal_digits": 90,
1641
+ "arithmetic": "high-precision numerical; not an exact certificate"
1642
+ }
1643
+ ],
1644
+ "exact_minimal_threshold": {
1645
+ "input_parameter": "1/2",
1646
+ "columns": [
1647
+ [
1648
+ 1,
1649
+ 0
1650
+ ],
1651
+ [
1652
+ 0,
1653
+ 1
1654
+ ],
1655
+ [
1656
+ 1,
1657
+ 64
1658
+ ],
1659
+ [
1660
+ 64,
1661
+ -1
1662
+ ]
1663
+ ],
1664
+ "input_root_product_lengths": [
1665
+ 4097,
1666
+ 4096,
1667
+ 1
1668
+ ],
1669
+ "normalized_output_products": [
1670
+ "2*sqrt(4097)",
1671
+ "81",
1672
+ "81"
1673
+ ],
1674
+ "output_root_product_lengths": [
1675
+ 16388,
1676
+ 6561,
1677
+ 6561
1678
+ ],
1679
+ "exact_output_root_defect": 3266,
1680
+ "input_support": "U_(2,4); all six minors nonzero",
1681
+ "arithmetic": "exact integers and stated algebraic square roots"
1682
+ },
1683
+ "exact_two_peak_weak": {
1684
+ "instances": [
1685
+ {
1686
+ "rank": 3,
1687
+ "ground_size": 6,
1688
+ "weak_relations": 30,
1689
+ "root_product_patterns": {
1690
+ "('1', '1', '2')": 18,
1691
+ "('1', '1', '1')": 12
1692
+ },
1693
+ "failed_strong_relations": 6,
1694
+ "arithmetic": "exact rational root products"
1695
+ },
1696
+ {
1697
+ "rank": 4,
1698
+ "ground_size": 7,
1699
+ "weak_relations": 105,
1700
+ "root_product_patterns": {
1701
+ "('1', '1', '1')": 69,
1702
+ "('1', '1', '2')": 36
1703
+ },
1704
+ "failed_strong_relations": 6,
1705
+ "arithmetic": "exact rational root products"
1706
+ },
1707
+ {
1708
+ "rank": 5,
1709
+ "ground_size": 8,
1710
+ "weak_relations": 280,
1711
+ "root_product_patterns": {
1712
+ "('1', '1', '1')": 220,
1713
+ "('1', '1', '2')": 60
1714
+ },
1715
+ "failed_strong_relations": 6,
1716
+ "arithmetic": "exact rational root products"
1717
+ },
1718
+ {
1719
+ "rank": 4,
1720
+ "ground_size": 8,
1721
+ "weak_relations": 420,
1722
+ "root_product_patterns": {
1723
+ "('1', '1', '1')": 348,
1724
+ "('1', '1', '2')": 72
1725
+ },
1726
+ "failed_strong_relations": 6,
1727
+ "arithmetic": "exact rational root products"
1728
+ }
1729
+ ],
1730
+ "universal_output_products": [
1731
+ "(1+2^q)^2",
1732
+ "4",
1733
+ "4"
1734
+ ],
1735
+ "necessary_output_parameter": "p >= 2*log_2((1+2^q)/2)",
1736
+ "qualification": "The universal formula requires the written symbolic proof."
1737
+ },
1738
+ "exact_two_peak_strong": {
1739
+ "instances": [
1740
+ {
1741
+ "rank": 3,
1742
+ "ground_size": 6,
1743
+ "weak_relation_count": 30,
1744
+ "strong_genuine_relation_count": 135,
1745
+ "strong_tautological_relation_count": 90,
1746
+ "strong_root_peak_values": [
1747
+ "2",
1748
+ "3/2"
1749
+ ],
1750
+ "strict_integer_coefficient_values": [
1751
+ 6,
1752
+ 14,
1753
+ 23
1754
+ ],
1755
+ "strict_integer_input_relations_certified": 135,
1756
+ "exact_minimum_root_margin_lower_bound": "2782077753681962262827255313504123441697893855430707882474822073622780887257257/50000000000000000000000000000000000000000000000000000000000000000000000000000000",
1757
+ "output_coefficients": {
1758
+ "12": "12",
1759
+ "13": "29",
1760
+ "14": "12",
1761
+ "23": "12",
1762
+ "24": "20",
1763
+ "34": "12"
1764
+ },
1765
+ "output_products": [
1766
+ 580,
1767
+ 144,
1768
+ 144
1769
+ ],
1770
+ "exact_squared_output_excess": 4,
1771
+ "arithmetic": "exact rational root products and rational square-root intervals"
1772
+ },
1773
+ {
1774
+ "rank": 4,
1775
+ "ground_size": 7,
1776
+ "weak_relation_count": 105,
1777
+ "strong_genuine_relation_count": 525,
1778
+ "strong_tautological_relation_count": 210,
1779
+ "strong_root_peak_values": [
1780
+ "2",
1781
+ "3/2"
1782
+ ],
1783
+ "strict_integer_coefficient_values": [
1784
+ 6,
1785
+ 14,
1786
+ 23
1787
+ ],
1788
+ "strict_integer_input_relations_certified": 525,
1789
+ "exact_minimum_root_margin_lower_bound": "2782077753681962262827255313504123441697893855430707882474822073622780887257257/50000000000000000000000000000000000000000000000000000000000000000000000000000000",
1790
+ "output_coefficients": {
1791
+ "12": "12",
1792
+ "13": "29",
1793
+ "14": "12",
1794
+ "23": "12",
1795
+ "24": "20",
1796
+ "34": "12"
1797
+ },
1798
+ "output_products": [
1799
+ 580,
1800
+ 144,
1801
+ 144
1802
+ ],
1803
+ "exact_squared_output_excess": 4,
1804
+ "arithmetic": "exact rational root products and rational square-root intervals"
1805
+ },
1806
+ {
1807
+ "rank": 5,
1808
+ "ground_size": 8,
1809
+ "weak_relation_count": 280,
1810
+ "strong_genuine_relation_count": 1540,
1811
+ "strong_tautological_relation_count": 420,
1812
+ "strong_root_peak_values": [
1813
+ "2",
1814
+ "3/2"
1815
+ ],
1816
+ "strict_integer_coefficient_values": [
1817
+ 6,
1818
+ 14,
1819
+ 23
1820
+ ],
1821
+ "strict_integer_input_relations_certified": 1540,
1822
+ "exact_minimum_root_margin_lower_bound": "2782077753681962262827255313504123441697893855430707882474822073622780887257257/50000000000000000000000000000000000000000000000000000000000000000000000000000000",
1823
+ "output_coefficients": {
1824
+ "12": "12",
1825
+ "13": "29",
1826
+ "14": "12",
1827
+ "23": "12",
1828
+ "24": "20",
1829
+ "34": "12"
1830
+ },
1831
+ "output_products": [
1832
+ 580,
1833
+ 144,
1834
+ 144
1835
+ ],
1836
+ "exact_squared_output_excess": 4,
1837
+ "arithmetic": "exact rational root products and rational square-root intervals"
1838
+ },
1839
+ {
1840
+ "rank": 4,
1841
+ "ground_size": 8,
1842
+ "weak_relation_count": 420,
1843
+ "strong_genuine_relation_count": 2576,
1844
+ "strong_tautological_relation_count": 560,
1845
+ "strong_root_peak_values": [
1846
+ "2",
1847
+ "3/2"
1848
+ ],
1849
+ "strict_integer_coefficient_values": [
1850
+ 6,
1851
+ 14,
1852
+ 23
1853
+ ],
1854
+ "strict_integer_input_relations_certified": 2576,
1855
+ "exact_minimum_root_margin_lower_bound": "2782077753681962262827255313504123441697893855430707882474822073622780887257257/50000000000000000000000000000000000000000000000000000000000000000000000000000000",
1856
+ "output_coefficients": {
1857
+ "12": "12",
1858
+ "13": "29",
1859
+ "14": "12",
1860
+ "23": "12",
1861
+ "24": "20",
1862
+ "34": "12"
1863
+ },
1864
+ "output_products": [
1865
+ 580,
1866
+ 144,
1867
+ 144
1868
+ ],
1869
+ "exact_squared_output_excess": 4,
1870
+ "arithmetic": "exact rational root products and rational square-root intervals"
1871
+ },
1872
+ {
1873
+ "rank": 5,
1874
+ "ground_size": 9,
1875
+ "weak_relation_count": 1260,
1876
+ "strong_genuine_relation_count": 9324,
1877
+ "strong_tautological_relation_count": 1260,
1878
+ "strong_root_peak_values": [
1879
+ "2",
1880
+ "3/2"
1881
+ ],
1882
+ "strict_integer_coefficient_values": [
1883
+ 6,
1884
+ 14,
1885
+ 23
1886
+ ],
1887
+ "strict_integer_input_relations_certified": 9324,
1888
+ "exact_minimum_root_margin_lower_bound": "2782077753681962262827255313504123441697893855430707882474822073622780887257257/50000000000000000000000000000000000000000000000000000000000000000000000000000000",
1889
+ "output_coefficients": {
1890
+ "12": "12",
1891
+ "13": "29",
1892
+ "14": "12",
1893
+ "23": "12",
1894
+ "24": "20",
1895
+ "34": "12"
1896
+ },
1897
+ "output_products": [
1898
+ 580,
1899
+ 144,
1900
+ 144
1901
+ ],
1902
+ "exact_squared_output_excess": 4,
1903
+ "arithmetic": "exact rational root products and rational square-root intervals"
1904
+ }
1905
+ ],
1906
+ "universal_two_peak_output_products": [
1907
+ "(1+2^q)*(1+(3/2)^q)",
1908
+ "4",
1909
+ "4"
1910
+ ],
1911
+ "necessary_output_parameter": "p >= log_2(((1+2^q)*(1+(3/2)^q))/4)",
1912
+ "exact_boundary_q2_output_products": [
1913
+ "65/4",
1914
+ "4",
1915
+ "4"
1916
+ ],
1917
+ "boundary_q2_necessary_output_parameter": "p >= log_2(65/16) > 2",
1918
+ "qualification": "Universal validity requires the written all-rank proof."
1919
+ },
1920
+ "exact_corank_relation_counts": {
1921
+ "arithmetic": "exact finite combinatorial enumeration",
1922
+ "instances": [
1923
+ {
1924
+ "rank": 2,
1925
+ "ground_size": 2,
1926
+ "corank": 0,
1927
+ "strong_term_counts": {},
1928
+ "maximum_summands": 0
1929
+ },
1930
+ {
1931
+ "rank": 2,
1932
+ "ground_size": 3,
1933
+ "corank": 1,
1934
+ "strong_term_counts": {
1935
+ "2": 3
1936
+ },
1937
+ "maximum_summands": 2
1938
+ },
1939
+ {
1940
+ "rank": 2,
1941
+ "ground_size": 4,
1942
+ "corank": 2,
1943
+ "strong_term_counts": {
1944
+ "2": 12,
1945
+ "3": 4
1946
+ },
1947
+ "maximum_summands": 3
1948
+ },
1949
+ {
1950
+ "rank": 2,
1951
+ "ground_size": 5,
1952
+ "corank": 3,
1953
+ "strong_term_counts": {
1954
+ "2": 30,
1955
+ "3": 20
1956
+ },
1957
+ "maximum_summands": 3
1958
+ },
1959
+ {
1960
+ "rank": 3,
1961
+ "ground_size": 3,
1962
+ "corank": 0,
1963
+ "strong_term_counts": {},
1964
+ "maximum_summands": 0
1965
+ },
1966
+ {
1967
+ "rank": 3,
1968
+ "ground_size": 4,
1969
+ "corank": 1,
1970
+ "strong_term_counts": {
1971
+ "2": 6
1972
+ },
1973
+ "maximum_summands": 2
1974
+ },
1975
+ {
1976
+ "rank": 3,
1977
+ "ground_size": 5,
1978
+ "corank": 2,
1979
+ "strong_term_counts": {
1980
+ "2": 30,
1981
+ "3": 20
1982
+ },
1983
+ "maximum_summands": 3
1984
+ },
1985
+ {
1986
+ "rank": 3,
1987
+ "ground_size": 6,
1988
+ "corank": 3,
1989
+ "strong_term_counts": {
1990
+ "2": 90,
1991
+ "3": 120,
1992
+ "4": 15
1993
+ },
1994
+ "maximum_summands": 4
1995
+ },
1996
+ {
1997
+ "rank": 4,
1998
+ "ground_size": 4,
1999
+ "corank": 0,
2000
+ "strong_term_counts": {},
2001
+ "maximum_summands": 0
2002
+ },
2003
+ {
2004
+ "rank": 4,
2005
+ "ground_size": 5,
2006
+ "corank": 1,
2007
+ "strong_term_counts": {
2008
+ "2": 10
2009
+ },
2010
+ "maximum_summands": 2
2011
+ },
2012
+ {
2013
+ "rank": 4,
2014
+ "ground_size": 6,
2015
+ "corank": 2,
2016
+ "strong_term_counts": {
2017
+ "2": 60,
2018
+ "3": 60
2019
+ },
2020
+ "maximum_summands": 3
2021
+ },
2022
+ {
2023
+ "rank": 4,
2024
+ "ground_size": 7,
2025
+ "corank": 3,
2026
+ "strong_term_counts": {
2027
+ "2": 210,
2028
+ "3": 420,
2029
+ "4": 105
2030
+ },
2031
+ "maximum_summands": 4
2032
+ },
2033
+ {
2034
+ "rank": 5,
2035
+ "ground_size": 5,
2036
+ "corank": 0,
2037
+ "strong_term_counts": {},
2038
+ "maximum_summands": 0
2039
+ },
2040
+ {
2041
+ "rank": 5,
2042
+ "ground_size": 6,
2043
+ "corank": 1,
2044
+ "strong_term_counts": {
2045
+ "2": 15
2046
+ },
2047
+ "maximum_summands": 2
2048
+ },
2049
+ {
2050
+ "rank": 5,
2051
+ "ground_size": 7,
2052
+ "corank": 2,
2053
+ "strong_term_counts": {
2054
+ "2": 105,
2055
+ "3": 140
2056
+ },
2057
+ "maximum_summands": 3
2058
+ },
2059
+ {
2060
+ "rank": 5,
2061
+ "ground_size": 8,
2062
+ "corank": 3,
2063
+ "strong_term_counts": {
2064
+ "2": 420,
2065
+ "3": 1120,
2066
+ "4": 420
2067
+ },
2068
+ "maximum_summands": 4
2069
+ },
2070
+ {
2071
+ "rank": 6,
2072
+ "ground_size": 6,
2073
+ "corank": 0,
2074
+ "strong_term_counts": {},
2075
+ "maximum_summands": 0
2076
+ },
2077
+ {
2078
+ "rank": 6,
2079
+ "ground_size": 7,
2080
+ "corank": 1,
2081
+ "strong_term_counts": {
2082
+ "2": 21
2083
+ },
2084
+ "maximum_summands": 2
2085
+ },
2086
+ {
2087
+ "rank": 6,
2088
+ "ground_size": 8,
2089
+ "corank": 2,
2090
+ "strong_term_counts": {
2091
+ "2": 168,
2092
+ "3": 280
2093
+ },
2094
+ "maximum_summands": 3
2095
+ },
2096
+ {
2097
+ "rank": 6,
2098
+ "ground_size": 9,
2099
+ "corank": 3,
2100
+ "strong_term_counts": {
2101
+ "2": 756,
2102
+ "3": 2520,
2103
+ "4": 1260
2104
+ },
2105
+ "maximum_summands": 4
2106
+ }
2107
+ ]
2108
+ }
2109
+ }
results/strong_parameter_bound_results.json ADDED
The diff for this file is too large to render. See raw diff
 
results/verification_results.json ADDED
The diff for this file is too large to render. See raw diff