Research map / Reinforcement learning / Policy gradient and actor-critic
Reward Redistribution for CVaR MDPs using a Bellman Operator on L-infinity
Aneri Muni, Vincent Taboga, Esther Derman, Pierre-Luc Bacon, Erick Delage
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 |
|---|---|---|---|
| Use the transformed CVaR Bellman operator with dense per-step reward instead of the sparse-reward augmented operator, on bounded functions | static CVaR value iteration · theory (bounded value functions) | fixed point / contraction | Improved |
| Increase discretization resolution of the augmented state (100 to 5000 bins) (ablation) | static CVaR value iteration with lower/upper bounding operators · stochastic gridworld | gap between upper and lower bound CVaR performance | Improved |
| Lower risk level alpha (toward robust) with the learned CVaR policies | static CVaR Q-VI and Q-learning · stochastic gridworld | lower tail of return distribution / crater entries | Improved |
| Use risk-averse CVaR policy instead of risk-neutral policy at small alpha | static CVaR agents · stochastic gridworld | empirical CVaR | Improved |
| Use model-free Q-learning instead of Q-VI | static CVaR algorithms · stochastic gridworld, alpha in [0.5, 0.6] | policy behavior and variance | Mixed |
| Run static CVaR Q-learning with block updates over all budgets for 50k iterations | static CVaR Q-learning · stochastic gridworld | distance to Q-VI value function | Improved |
Closest papers by meaning
- Provably Efficient CVaR RL in Low-rank MDPs (ICLR 2024 · similarity 0.89)
- A Reductions Approach to Risk-Sensitive Reinforcement Learning with Optimized Certainty Equivalents (ICML 2025 · similarity 0.88)
- Risk-Sensitive Policy Optimization via Predictive CVaR Policy Gradient (ICML 2024 · similarity 0.88)
- Risk-Averse Online POMDP Planning via CVaR of the Immediate Cost with Performance Guarantees (NeurIPS 2026 · similarity 0.88)
- Risk-Sensitive Reward-Free Reinforcement Learning with CVaR (ICML 2024 · similarity 0.87)
- Planning and Learning in Average Risk-aware MDPs (NeurIPS 2025 · 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.
