{ "contract_source_sha": "498fa31f672759981d3039c03705cb720cd0ac799f78fd78e33acbcc71e5ee18", "steps": 64, "results": [ { "seed": 17, "arm": "stateless", "validation": { "W3_MSE": 0.11103641552229722, "W4_MSE": 0.10791260438660781, "ctrl_max_rms": 0.00528129100954781 }, "checkpoint_step": 64 }, { "seed": 17, "arm": "mhc", "validation": { "W3_MSE": 0.11103636150558789, "W4_MSE": 0.10791248766084512, "ctrl_max_rms": 0.005226058148158093 }, "checkpoint_step": 64 }, { "seed": 17, "arm": "mhc_noaux", "validation": { "W3_MSE": 0.1110341747601827, "W4_MSE": 0.10790904238820076, "ctrl_max_rms": 0.0052098391655211644 }, "checkpoint_step": 64 } ], "history": [ { "seed": 17, "arm": "stateless", "step": 0, "loss_train": 0.09119217842817307, "grad_norm": 0.013730722479522228, "head_grad_norm": 1.4221210449250066e-06, "q0_trials": [ { "radius": 0.0003, "relative_decrease": 4.501778529326963e-05, "max_action": 4.158697830547453e-09 }, { "radius": 0.0001, "relative_decrease": 1.503316242098296e-05, "max_action": 1.3862333503311675e-09 }, { "radius": 3e-05, "relative_decrease": 4.411906362679782e-06, "max_action": 4.158700883660771e-10 }, { "radius": 1e-05, "relative_decrease": 1.470635454226594e-06, "max_action": 3.463760112065728e-10 } ], "W3_MSE": 0.11472809066375096, "W4_MSE": 0.11399433637658755, "ctrl_max_rms": 3.1482448361227117e-10 }, { "seed": 17, "arm": "stateless", "step": 8, "loss_train": 0.08606645464897156, "W3_MSE": 0.1147225337723891, "W4_MSE": 0.11396808549761772, "ctrl_max_rms": 0.00018752604228211567 }, { "seed": 17, "arm": "stateless", "step": 16, "loss_train": 0.1242414116859436, "W3_MSE": 0.11465953538815181, "W4_MSE": 0.11387925470868747, "ctrl_max_rms": 0.00030438662846184644 }, { "seed": 17, "arm": "stateless", "step": 24, "loss_train": 0.09662917256355286, "W3_MSE": 0.11453648408253987, "W4_MSE": 0.11368757920960586, "ctrl_max_rms": 0.0003923155600205064 }, { "seed": 17, "arm": "stateless", "step": 32, "loss_train": 0.1582513302564621, "W3_MSE": 0.1143090259283781, "W4_MSE": 0.11331315711140633, "ctrl_max_rms": 0.0005296767340041697 }, { "seed": 17, "arm": "stateless", "step": 40, "loss_train": 0.10655954480171204, "W3_MSE": 0.11380417520801227, "W4_MSE": 0.11249382669727008, "ctrl_max_rms": 0.000945325159894613 }, { "seed": 17, "arm": "stateless", "step": 48, "loss_train": 0.10963404178619385, "W3_MSE": 0.1130196029941241, "W4_MSE": 0.11111373826861382, "ctrl_max_rms": 0.00205805353471078 }, { "seed": 17, "arm": "stateless", "step": 56, "loss_train": 0.10562591254711151, "W3_MSE": 0.11191886352996032, "W4_MSE": 0.10920579296847184, "ctrl_max_rms": 0.004299115234365066 }, { "seed": 17, "arm": "stateless", "step": 64, "loss_train": 0.1330445259809494, "W3_MSE": 0.11103641552229722, "W4_MSE": 0.10791260438660781, "ctrl_max_rms": 0.00528129100954781 }, { "seed": 17, "arm": "mhc", "step": 0, "loss_train": 0.09119217842817307, "grad_norm": 0.013730722479522228, "head_grad_norm": 1.4206758578438894e-06, "q0_trials": [ { "radius": 0.0003, "relative_decrease": 4.501778529326963e-05, "max_action": 4.1586671883919735e-09 }, { "radius": 0.0001, "relative_decrease": 1.503316242098296e-05, "max_action": 1.3862230252570384e-09 }, { "radius": 3e-05, "relative_decrease": 4.411906362679782e-06, "max_action": 4.1586703525275936e-10 }, { "radius": 1e-05, "relative_decrease": 1.470635454226594e-06, "max_action": 3.463760112065728e-10 } ], "W3_MSE": 0.11472809066375096, "W4_MSE": 0.11399433637658755, "ctrl_max_rms": 3.1482448361227117e-10 }, { "seed": 17, "arm": "mhc", "step": 8, "loss_train": 0.08606645464897156, "W3_MSE": 0.11472253315150738, "W4_MSE": 0.113968084876736, "ctrl_max_rms": 0.00018755901813468276 }, { "seed": 17, "arm": "mhc", "step": 16, "loss_train": 0.1242414116859436, "W3_MSE": 0.11465953538815181, "W4_MSE": 0.11387925657133262, "ctrl_max_rms": 0.0003044217810383998 }, { "seed": 17, "arm": "mhc", "step": 24, "loss_train": 0.09662917256355286, "W3_MSE": 0.11453648408253987, "W4_MSE": 0.11368757796784242, "ctrl_max_rms": 0.00039232581184478477 }, { "seed": 17, "arm": "mhc", "step": 32, "loss_train": 0.1582513302564621, "W3_MSE": 0.11430902406573296, "W4_MSE": 0.11331315897405148, "ctrl_max_rms": 0.0005288315539170677 }, { "seed": 17, "arm": "mhc", "step": 40, "loss_train": 0.10655954480171204, "W3_MSE": 0.11380419631799062, "W4_MSE": 0.11249386705458164, "ctrl_max_rms": 0.0009341221496773263 }, { "seed": 17, "arm": "mhc", "step": 48, "loss_train": 0.109634168446064, "W3_MSE": 0.11301966508229573, "W4_MSE": 0.11111390093962352, "ctrl_max_rms": 0.0020173621790794036 }, { "seed": 17, "arm": "mhc", "step": 56, "loss_train": 0.10562613606452942, "W3_MSE": 0.1119188933322827, "W4_MSE": 0.10920589851836364, "ctrl_max_rms": 0.004241040248113374 }, { "seed": 17, "arm": "mhc", "step": 64, "loss_train": 0.13304445147514343, "W3_MSE": 0.11103636150558789, "W4_MSE": 0.10791248766084512, "ctrl_max_rms": 0.005226058148158093 }, { "seed": 17, "arm": "mhc_noaux", "step": 0, "loss_train": 0.08952551335096359, "grad_norm": 0.01346376072615385, "head_grad_norm": 1.393729007759248e-06, "q0_trials": [ { "radius": 0.0003, "relative_decrease": 4.502363574430603e-05, "max_action": 4.158414501631569e-09 }, { "radius": 0.0001, "relative_decrease": 1.5063360572494254e-05, "max_action": 1.3861388703517719e-09 }, { "radius": 3e-05, "relative_decrease": 4.4108182891834e-06, "max_action": 4.1584174992337353e-10 }, { "radius": 1e-05, "relative_decrease": 1.4980137585905887e-06, "max_action": 3.463760112065728e-10 } ], "W3_MSE": 0.11472809066375096, "W4_MSE": 0.11399433637658755, "ctrl_max_rms": 3.1482448361227117e-10 }, { "seed": 17, "arm": "mhc_noaux", "step": 8, "loss_train": 0.08451367169618607, "W3_MSE": 0.11472250955800216, "W4_MSE": 0.11396805879970391, "ctrl_max_rms": 0.000187463211962798 }, { "seed": 17, "arm": "mhc_noaux", "step": 16, "loss_train": 0.12192638218402863, "W3_MSE": 0.11465946398675442, "W4_MSE": 0.11387916964789231, "ctrl_max_rms": 0.00030426468583755195 }, { "seed": 17, "arm": "mhc_noaux", "step": 24, "loss_train": 0.09486307948827744, "W3_MSE": 0.11453630837301414, "W4_MSE": 0.1136873426536719, "ctrl_max_rms": 0.00039202869569029036 }, { "seed": 17, "arm": "mhc_noaux", "step": 32, "loss_train": 0.1553032398223877, "W3_MSE": 0.11430866395433743, "W4_MSE": 0.11331262377401193, "ctrl_max_rms": 0.0005277997018614163 }, { "seed": 17, "arm": "mhc_noaux", "step": 40, "loss_train": 0.10460616648197174, "W3_MSE": 0.11380349472165108, "W4_MSE": 0.1124927494674921, "ctrl_max_rms": 0.0009292569302488118 }, { "seed": 17, "arm": "mhc_noaux", "step": 48, "loss_train": 0.10761874914169312, "W3_MSE": 0.11301849285761516, "W4_MSE": 0.11111204698681831, "ctrl_max_rms": 0.002005132545794671 }, { "seed": 17, "arm": "mhc_noaux", "step": 56, "loss_train": 0.10368967801332474, "W3_MSE": 0.11191730387508869, "W4_MSE": 0.10920340940356255, "ctrl_max_rms": 0.004224800640561928 }, { "seed": 17, "arm": "mhc_noaux", "step": 64, "loss_train": 0.1305774450302124, "W3_MSE": 0.1110341747601827, "W4_MSE": 0.10790904238820076, "ctrl_max_rms": 0.0052098391655211644 }, { "seed": 41, "arm": "stateless", "step": 0, "loss_train": 0.12777335941791534, "grad_norm": 0.006554036866873503, "head_grad_norm": 1.771895426827541e-06, "q0_trials": [ { "radius": 0.0003, "relative_decrease": 1.5277457877657884e-05, "max_action": 2.5834463457385937e-09 }, { "radius": 0.0001, "relative_decrease": 5.014738081979306e-06, "max_action": 8.611487078979962e-10 }, { "radius": 3e-05, "relative_decrease": 1.51608360617979e-06, "max_action": 5.335194508404584e-10 }, { "radius": 1e-05, "relative_decrease": 4.664872634399354e-07, "max_action": 5.335235031544983e-10 } ], "W3_MSE": 0.11472141183912754, "W4_MSE": 0.11397158540785313, "ctrl_max_rms": 4.782619277462116e-10 }, { "seed": 41, "arm": "stateless", "step": 8, "loss_train": 0.10454776883125305, "W3_MSE": 0.11467163388927777, "W4_MSE": 0.11390550496677558, "ctrl_max_rms": 9.996155737705219e-05 }, { "seed": 41, "arm": "stateless", "step": 16, "loss_train": 0.09654892981052399, "W3_MSE": 0.11463494785130024, "W4_MSE": 0.1138407854984204, "ctrl_max_rms": 0.00016169367881957442 }, { "seed": 41, "arm": "stateless", "step": 24, "loss_train": 0.11449190974235535, "W3_MSE": 0.11451686484118302, "W4_MSE": 0.11365282163023949, "ctrl_max_rms": 0.00030108772868212935 } ], "checkpoints": [ { "sha256": "488db3159c0ffdedf1c4df4b40875a86857a66cd26b6469c53b89cebf2e776ba", "hf_path": "experiments/MHC_REAL_A1/pilot_seed17_stateless_step0000.pt", "bytes": 93186 }, { "sha256": "8564fa6b02b9004df33b139b15888e345f27bc9392db5b1bdd36918417abb5d3", "hf_path": "experiments/MHC_REAL_A1/pilot_seed17_stateless_step0008.pt", "bytes": 268976 }, { "sha256": "42792020756dbc501ab231c407134a90b13db10de7d9dca0192c4da6eabafbae", "hf_path": "experiments/MHC_REAL_A1/pilot_seed17_stateless_step0016.pt", "bytes": 268976 }, { "sha256": "c1297071cf2cd8ef2da147b116cd01823b41f9af63bc2a603e075a72ec9936ca", "hf_path": "experiments/MHC_REAL_A1/pilot_seed17_stateless_step0024.pt", "bytes": 268976 }, { "sha256": "c95fcb3a196dfed881e2a01f5b54b8dea25fbfe13f5eae71336af1c671f71d14", "hf_path": "experiments/MHC_REAL_A1/pilot_seed17_stateless_step0032.pt", "bytes": 268976 }, { "sha256": "84d1bb51dafe0dc9ab687d2bfaa6f2e590a6a257a72053a9d05610754a27e554", "hf_path": "experiments/MHC_REAL_A1/pilot_seed17_stateless_step0040.pt", "bytes": 268976 }, { "sha256": "a2a4ffa22bf4208049e3b06accfb87699cb28b0987592bbbbdada6aea7410750", "hf_path": "experiments/MHC_REAL_A1/pilot_seed17_stateless_step0048.pt", "bytes": 268976 }, { "sha256": "99d7cae2f26351d8183ceb0b2c152adf512324e147ffbce0c8b0a5fd4c34163a", "hf_path": "experiments/MHC_REAL_A1/pilot_seed17_stateless_step0056.pt", "bytes": 268976 }, { "sha256": "e2a449c3025fe0223022be26a6a9c1ee8e6a60e46f146208f6a7a532f9dba37e", "hf_path": "experiments/MHC_REAL_A1/pilot_seed17_stateless_step0064.pt", "bytes": 268976 }, { "sha256": "13adfa1b732166cc1eec631d7cbbe5cac3e25ffe4d0f75061a078bd37f6f695b", "hf_path": "experiments/MHC_REAL_A1/pilot_seed17_mhc_step0000.pt", "bytes": 92972 }, { "sha256": "b6f4373dc8e2fe998899539e0230b7ace5b19586f2a915be4525631cfdcaf1e4", "hf_path": "experiments/MHC_REAL_A1/pilot_seed17_mhc_step0008.pt", "bytes": 268502 }, { "sha256": "31dd6dfa2ca3fc25a4208d0bf48363fa20b901f95deda844947dd89b40f5d4b4", "hf_path": "experiments/MHC_REAL_A1/pilot_seed17_mhc_step0016.pt", "bytes": 268502 }, { "sha256": "32fd7185145f7e049b8b4aaf4b92379781f4176abf96d8a03d388c56a1df644c", "hf_path": "experiments/MHC_REAL_A1/pilot_seed17_mhc_step0024.pt", "bytes": 268502 }, { "sha256": "72b6f923cafa30228fab4697778d1b4df557b854ad86c4114f519a2cf5b84c22", "hf_path": "experiments/MHC_REAL_A1/pilot_seed17_mhc_step0032.pt", "bytes": 268502 }, { "sha256": "68bfa8209125c401c1a11031e3c95e8bcd15181e1ba12f4fecb743f33cfd038d", "hf_path": "experiments/MHC_REAL_A1/pilot_seed17_mhc_step0040.pt", "bytes": 268502 }, { "sha256": "ee57d9014f55ead2d39be0a8aa54d004516a852b83252eb2f3d73d7bcc6f3925", "hf_path": "experiments/MHC_REAL_A1/pilot_seed17_mhc_step0048.pt", "bytes": 268502 }, { "sha256": "58ab07249ff1526f54f3d26304c77666e186393a8558e698d8b6a14179a2575f", "hf_path": "experiments/MHC_REAL_A1/pilot_seed17_mhc_step0056.pt", "bytes": 268502 }, { "sha256": "e55f47fb3ecd1f1518e0ec629d1f24311b2e05f6a8fd07fe04ac7de1948528d6", "hf_path": "experiments/MHC_REAL_A1/pilot_seed17_mhc_step0064.pt", "bytes": 268502 }, { "sha256": "bf78c84f9f1dcb58d3d5f604b3dd66703ea23496c74da29c549c5c2ece4eb97f", "hf_path": "experiments/MHC_REAL_A1/pilot_seed17_mhc_noaux_step0000.pt", "bytes": 93186 }, { "sha256": "3d4aec2173d08053ae9800cd45bf53458c28c552c60ebd85ddc055a9bd5a60e2", "hf_path": "experiments/MHC_REAL_A1/pilot_seed17_mhc_noaux_step0008.pt", "bytes": 268976 }, { "sha256": "a78f7743de7696e69afc74f8df091af9b1cadd68ac7660e56d232207263817b2", "hf_path": "experiments/MHC_REAL_A1/pilot_seed17_mhc_noaux_step0016.pt", "bytes": 268976 }, { "sha256": "e9165957b3f41da8f0e0fe6f66b823b002ac3ea5cdaf3f0d509a426ec2696119", "hf_path": "experiments/MHC_REAL_A1/pilot_seed17_mhc_noaux_step0024.pt", "bytes": 268976 }, { "sha256": "4acbd87fb99ac2eaad7ebb1070fbb36336346ca22dd8f39851a0e424ec7ae77e", "hf_path": "experiments/MHC_REAL_A1/pilot_seed17_mhc_noaux_step0032.pt", "bytes": 268976 }, { "sha256": "a19656d8a65e013ac5594640064407da33bb7f6a728f073a27b438651a1a48a8", "hf_path": "experiments/MHC_REAL_A1/pilot_seed17_mhc_noaux_step0040.pt", "bytes": 268976 }, { "sha256": "ee9d7a4757ffd0fb173d317bb6ac1f9a82c0c336525eaf9a1ddc196bc4abb5c6", "hf_path": "experiments/MHC_REAL_A1/pilot_seed17_mhc_noaux_step0048.pt", "bytes": 268976 }, { "sha256": "e6ee9bf366d2c7204b4978e0108c6aaa27e804a03f261d6b8ccdf65dc8a94006", "hf_path": "experiments/MHC_REAL_A1/pilot_seed17_mhc_noaux_step0056.pt", "bytes": 268976 }, { "sha256": "0031e8e356e41310dbde85f7db8ac475b27328a20b661e971d6b548fedc95b91", "hf_path": "experiments/MHC_REAL_A1/pilot_seed17_mhc_noaux_step0064.pt", "bytes": 268976 }, { "sha256": "f77e831f0151146343715cca301e0dc76183d5f1f63b30cd2b98a33bab11b0d1", "hf_path": "experiments/MHC_REAL_A1/pilot_seed41_stateless_step0000.pt", "bytes": 93186 }, { "sha256": "d934489f5dcecd5ae7d831b52e1ca108c4866f2028f8b338745d3986bc4f8cd6", "hf_path": "experiments/MHC_REAL_A1/pilot_seed41_stateless_step0008.pt", "bytes": 268976 }, { "sha256": "e5d497263f3240df7bc9f8984dfdaea0c41a12763190413868ebf92da29fc2af", "hf_path": "experiments/MHC_REAL_A1/pilot_seed41_stateless_step0016.pt", "bytes": 268976 }, { "sha256": "343f6069d3856e209c727c1aae123baf4e76e616b5988dcd27ed6ada99a2e54e", "hf_path": "experiments/MHC_REAL_A1/pilot_seed41_stateless_step0024.pt", "bytes": 268976 } ] }