Spaces:
Configuration error
Configuration error
|
Download docs/environment.md from Scikitlearn10801/DataPrivacyEnv: direct link, hf CLI and curl.
- Browser
- Download file 1.38 kB
-
https://huggingface.co/spaces/Scikitlearn10801/DataPrivacyEnv/resolve/main/docs/environment.md
- Command line
-
hf download hf://spaces/Scikitlearn10801/DataPrivacyEnv/docs/environment.md
-
curl -L -o environment.md https://huggingface.co/spaces/Scikitlearn10801/DataPrivacyEnv/resolve/main/docs/environment.md
1.38 kB
π§© Environment Specification
DataPrivacyEnv follows the Markov Decision Process (MDP) framework where an agent learns to minimize privacy risk through interaction.
π’ State Space ($S$)
The state is represented as an in-memory relational database containing:
- Users Table: User metadata (ID, Name, Role, and potentially exposed PII like passwords).
- Logs Table: Unstructured text data containing system events and leaked identifiers.
ποΈ Observation Space ($O$)
The agent does not see the full database state at once. It perceives:
status_code: The result of its last API action.response_data: The specific JSON fragment returned by the endpoint.current_task_goal: A natural language embedding of the objective.
β‘ Action Space ($A$)
The action space is discrete-symbolic. The agent must generate a valid PrivacyAction object:
- Methods:
GET,PATCH,DELETE. - Endpoints: Targeted resource paths.
- Payload: Key-value pairs for redaction logic.
π° Reward Function ($R$)
The reward is calculated based on state-delta verification:
- Success: $1.0$ (Task completed).
- Partial: Scaled based on percentage of PII fixed (e.g., $0.33$ per user record cleaned).
- Penalty: $0.0$ for invalid endpoints or malformed JSON.