Research map / Reinforcement learning / Policy gradient and actor-critic
Policy-Guided Heuristic Search with Guarantees
Laurent Orseau, Levi H. S. Lelis
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 PHS* instead of PHS_h (ablation) | PHS · Sokoban | expansions: 2,130.4 -> 1,522.1 | Improved |
| Use PHS* instead of PHS_h (ablation) | PHS · The Witness | expansions: 408.1 -> 781.5 | Hurt |
| Use PHS* instead of PHS_h (ablation) | PHS · Sliding Tile Puzzle | expansions: 58,692.0 -> 2,867.2 | Improved |
| Add a learned heuristic to policy-guided LevinTS search (PHS_h) (ablation) | LevinTS · Sokoban | expansions: 2,640.4 -> 2,130.4 | Improved |
| Add a learned heuristic to policy-guided LevinTS search (PHS_h) (ablation) | LevinTS · The Witness | expansions: 520.2 -> 408.1 | Improved |
| Add the admissible A*-style heuristic factor rather than using policy guidance alone (ablation) | LevinTS · Sliding Tile Puzzle | expansions: 39,005.6 -> 58,692.0 | Hurt |
| Use policy guidance with the heuristic rather than heuristic guidance alone | PHS_h · Sokoban | expansions: 8,696.8 -> 2,130.4 | Improved |
| Use policy guidance with the heuristic rather than heuristic guidance alone | PHS_h · The Witness | expansions: 9,022.3 -> 408.1 | Improved |
Closest papers by meaning
- Subgoal-Guided Policy Heuristic Search with Learned Subgoals (ICML 2025 · similarity 0.87)
- Robust and Diverse Multi-Agent Learning via Rational Policy Gradient (NeurIPS 2025 · similarity 0.86)
- Going Beyond Heuristics by Imposing Policy Improvement as a Constraint (NeurIPS 2024 · similarity 0.85)
- DGPO: Discovering Multiple Strategies with Diversity-Guided Policy Optimization (AAAI 2024 · similarity 0.84)
- Adversarial Policy Optimization for Offline Preference-based Reinforcement Learning (ICLR 2025 · similarity 0.84)
- Policy-Regret Minimization in Markov Games with Function Approximation (ICML 2025 · similarity 0.84)
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.
