Research map / Reinforcement learning / MDPs and planning
Risk-Averse Bayes-Adaptive Reinforcement Learning
Marc Rigter, Bruno Lacerda, Nick Hawes
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| optimize CVaR in the Bayes-adaptive MDP with RA-BAMCP rather than expected value with BAMCP | Betting Game · Betting Game | CVaR (α=0.03): 0.00 (0.0) -> 9.98 (0.02) | Improved |
| optimize CVaR in the Bayes-adaptive MDP with RA-BAMCP rather than expected value with BAMCP | Autonomous Car Navigation · Autonomous Car Navigation | CVaR (α=0.03): 5.56 (0.60) -> 23.51 (0.17) | Improved |
| optimize CVaR rather than expected value | RA-BAMCP on Autonomous Car Navigation · Autonomous Car Navigation | Expected value: 50.16 (0.38) -> 38.33 (0.15) | Hurt |
| optimize CVaR in the Bayes-adaptive MDP rather than use CVaR VI on the expected MDP | Betting Game · Betting Game | CVaR (α=0.03): 0.00 (0.0) -> 9.98 (0.02) | Improved |
| optimize CVaR in the Bayes-adaptive MDP rather than use CVaR VI on the expected MDP | Autonomous Car Navigation · Autonomous Car Navigation | CVaR (α=0.2): 20.42 (0.44) -> 27.76 (0.24) | Improved |
| use Bayesian optimization instead of random expansion for adversary actions (ablation) | RA-BAMCP · Betting Game | CVaR (α=0.03): 8.24 (0.29) -> 9.98 (0.02) | Improved |
| use Bayesian optimization instead of random expansion for adversary actions (ablation) | RA-BAMCP · Betting Game | CVaR (α=0.2): 18.39 (0.99) -> 20.09 (1.04) | Improved |
| use Bayesian optimization instead of random expansion for adversary actions (ablation) | RA-BAMCP · Autonomous Car Navigation | CVaR (α=0.03): 16.06 (0.35) -> 23.51 (0.17) | Improved |
| use Bayesian optimization instead of random expansion for adversary actions (ablation) | RA-BAMCP · Autonomous Car Navigation | CVaR (α=0.2): 26.84 (0.38) -> 27.76 (0.24) | Improved |
| lower the CVaR confidence level from α=0.2 to α=0.03 | RA-BAMCP on Autonomous Car Navigation · Autonomous Car Navigation | Expected value: 38.33 (0.15) -> 29.56 (0.06) | Hurt |
Closest papers by meaning
- Risk-Averse Total-Reward Reinforcement Learning (NeurIPS 2025 · similarity 0.88)
- A Reductions Approach to Risk-Sensitive Reinforcement Learning with Optimized Certainty Equivalents (ICML 2025 · similarity 0.88)
- Bayesian Exploration Networks (ICML 2024 · similarity 0.88)
- Pessimism Meets Risk: Risk-Sensitive Offline Reinforcement Learning (ICML 2024 · similarity 0.87)
- Risk-averse Total-reward MDPs with ERM and EVaR (AAAI 2025 · similarity 0.87)
- Efficient Exploration in Average-Reward Constrained Reinforcement Learning: Achieving Near-Optimal Regret With Posterior Sampling (ICML 2024 · similarity 0.87)
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.
