TEP-Dreamer Implementation Skill
CommunitySafe DreamerV3 for Tennessee Eastman process.
AuthorKK1182112KK
Version1.0.0
Installs0
System Documentation
What problem does it solve?
This Skill guides implementing a DreamerV3-based world-model reinforcement learning system with a risk-bounded safety shield for the Tennessee Eastman Process (TEP).
Core Features & Use Cases
- DreamerV3 World Model: Learn latent dynamics and perform imagined rollouts to enable planning and control.
- Constraint Head with Quantile Regression: Predicts physical constraint margins with calibrated uncertainty to guide safe interventions.
- Risk-Bounded Safety Shield: Quadratic/convex optimization-based action correction that minimally alters RL actions to satisfy safety budgets under a two-tier constraint scheme (Tier-1 trip and Tier-2 alarms).
- 4-Axis Holdout Evaluation: Tests generalization across disturbance type, intensity, transition, and compound disturbances.
- Phase-Driven Development: Structured progression from environment wiring to shield integration and evaluation.
- Visualization & Dashboard (Nice-to-have): Optional live dashboards for monitoring margins and shield activity.
Quick Start
Follow the Quick Start steps to train and evaluate the Dreamer-based safe controller on the Tennessee Eastman Process.
Dependency Matrix
Required Modules
torchnumpyscipygymnasiumcvxpyosqphydra-coreomegaconfwandbpandasmatplotlibseabornplotlytqdmtensorboardpyyaml
Components
scripts
💻 Claude Code Installation
Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.
Please help me install this Skill: Name: TEP-Dreamer Implementation Skill Download link: https://github.com/KK1182112KK/tep-dreamer/archive/main.zip#tep-dreamer-implementation-skill Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
Agent Skills Search Helper
Install a tiny helper to your Agent, search and equip skill from 223,000+ vetted skills library on demand.