Research map / Reinforcement learning / MDPs and planning
Online Robust Planning Under Model Uncertainty: A Sample-Based Approach
Tamir Shazman, Idan Lev-Yehudi, Ron Benchetrit, Vadim Indelman
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 robust backups (RSS) instead of nominal Sparse Sampling under misspecified model | online planning · FrozenLake 8x8, rho 0.1 to 0.6 | average discounted return: 0.080 -> 0.118 at rho=0.6 | Improved |
| use RSS instead of SS when rho is small | online planning · FrozenLake 8x8, rho=0.1 | average discounted return: 0.172 -> 0.177 | No clear effect |
| use RSS instead of SS under increasing hazard-zone noise | online planning · CartPole, sigma_high 0.07 to 0.15 | discounted return and success rate | Improved |
| use RSS instead of SS in low-noise regime | online planning · CartPole low noise | discounted return | Hurt |
| give SS access to the true dynamics (ablation) | Sparse Sampling · FrozenLake 8x8 | average discounted return: 0.249 vs 0.177 (RSS, rho=0.1) | Improved |
| robust SAA backup adds a log factor to per-node cost | Sparse Sampling · complexity analysis | computational complexity | No clear effect |
Closest papers by meaning
- Graph Sparse Sampling: Breaking the Curse of the Horizon in Continuous MDP Planning (NeurIPS 2026 · similarity 0.91)
- Online Robust Reinforcement Learning Through Monte-Carlo Planning (ICML 2025 · similarity 0.88)
- Factored Online Planning in Many-Agent POMDPs (AAAI 2024 · similarity 0.88)
- Near-Optimal Distributionally Robust Reinforcement Learning with General $L_p$ Norms (NeurIPS 2024 · similarity 0.88)
- Solving Robust Markov Decision Processes: Generic, Reliable, Efficient (AAAI 2025 · similarity 0.88)
- Distributionally Robust Reinforcement Learning with Interactive Data Collection: Fundamental Hardness and Near-Optimal Algorithms (NeurIPS 2024 · similarity 0.88)
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.
