Research map / Reinforcement learning / Policy gradient and actor-critic
How to Combine Tree-Search Methods in Reinforcement Learning
Yonathan Efroni, Gal Dalal, Bruno Scherrer, Shie Mannor
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 |
|---|---|---|---|
| Replace backing up v with backing up T^{h-1}v (ablation) | Partial-evaluation h-PI · Deterministic grid world, noiseless | simulator calls to convergence | Improved |
| Replace backing up v with backing up T^{h-1}v (ablation) | Partial-evaluation h-PI · Deterministic grid world, evaluation noise | ||v*-v^{π_f}||∞ | Improved |
| Increase h (ablation) | NC-hm-PI and hm-PI · Deterministic grid world, evaluation noise | ||v*-v^{π_f}||∞ | Improved |
| Increase m (ablation) | NC-hm-PI relative to hm-PI · Deterministic grid world, noiseless | difference in simulator calls to convergence | No clear effect |
| Set h=1, making the backup variants equivalent (ablation) | NC-hm-PI and hm-PI · Deterministic grid world, noiseless | simulator calls to convergence | No clear effect |
| Use v as the evaluation bootstrap instead of T^{h-1}v (ablation) | h-greedy partial-evaluation PI · General discounted MDP | contraction coefficient: γ^h -> γ^m+γ^h | Hurt |
| Increase h (ablation) | NC-hm-PI and hm-PI · Deterministic grid world, evaluation noise | sensitivity to approximation error | Mixed |
| Increase h (ablation) | hm-PI · Deterministic grid world, noiseless | best choice of m | Mixed |
Closest papers by meaning
- SPOT: Scalable Policy Optimization with Trees for Markov Decision Processes (NeurIPS 2025 · similarity 0.87)
- Reinforcement Learning with Multi-Step Lookahead Information Via Adaptive Batching (NeurIPS 2026 · similarity 0.87)
- Reinforcement Learning with Lookahead Information (NeurIPS 2024 · similarity 0.86)
- Monte-Carlo Tree Search with Uncertainty Propagation via Optimal Transport (ICML 2025 · similarity 0.86)
- Policy Gradient with Tree Expansion (ICML 2025 · similarity 0.86)
- Subgoal-Guided Policy Heuristic Search with Learned Subgoals (ICML 2025 · 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.
