Research map / Reinforcement learning / Policy gradient and actor-critic
Safe Policy Improvement by Minimizing Robust Baseline Regret
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| minimize robust baseline regret (RBC) instead of plain robust formulation | safe policy improvement · Illustrative example | return of optimal policy vs baseline: -9 - (-10) = 1 | Helped |
| use RBC heuristic instead of nominal model (EXP) | RBC · Example grid problem | improvement in return over baseline | Helped |
| use RBC instead of ROB and RWA with few samples | RBC · Example grid problem | improvement in return over baseline | Helped |
| reward-adjusted formulation (RWA) instead of robust | RWA · Illustrative example | safe improvement | Hurt |
| use RBC heuristic | RBC · Energy arbitrage | percentage improvement over baseline | Helped |
| increase number of transition samples (smaller model error) | all methods · Example grid problem | improvement over baseline | Helped |
Closest papers by meaning
- Accelerating Safe Reinforcement Learning with Constraint-mismatched Baseline Policies (ICML 2021 · similarity 0.88)
- Long-Term Safe Reinforcement Learning with Binary Feedback (AAAI 2024 · similarity 0.87)
- Pessimism Meets Risk: Risk-Sensitive Offline Reinforcement Learning (ICML 2024 · similarity 0.87)
- Flipping-based Policy for Chance-Constrained Markov Decision Processes (NeurIPS 2024 · similarity 0.86)
- Safe Reinforcement Learning with Instantaneous Constraints: The Role of Aggressive Exploration (AAAI 2024 · similarity 0.86)
- Off-Policy Primal-Dual Safe Reinforcement Learning (ICLR 2024 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.