Research map / Reinforcement learning / Policy gradient and actor-critic
Monte-Carlo Tree Search as Regularized Policy Optimization
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 MuZero's visit distribution with the regularized policy solution for acting, search, and learning | MuZero · Ms Pacman (Atari), 5 simulations | episodic return | Helped |
| Replace MuZero's visit distribution with the regularized policy solution for acting, search, and learning | MuZero · Ms Pacman (Atari), 50 simulations | episodic return | Helped |
| Replace MuZero's visit distribution with the regularized policy solution for acting, search, and learning | MuZero · Atari games, 50 simulations | episodic return | Helped |
| Replace MuZero's visit distribution with the regularized policy solution for acting, search, and learning | MuZero · Cheetah Run (Control Suite), low simulation budgets | episodic return | Helped |
| Use the regularized policy solution as the learning target (ablation) | MuZero · Ms Pacman (Atari), 5 simulations | episodic return | Helped |
| Use the regularized policy solution during search (ablation) | MuZero · Ms Pacman (Atari), 5 simulations | episodic return | Helped |
| Use the regularized policy solution for acting (ablation) | MuZero · Ms Pacman (Atari), 5 simulations | episodic return | Helped |
| Use the regularized policy solution as the learning target (ablation) | MuZero · Ms Pacman (Atari), 50 simulations | episodic return | No clear effect |
| Use the regularized policy solution for acting (ablation) | MuZero · Ms Pacman (Atari), 50 simulations | episodic return | No clear effect |
| Use the regularized policy solution during search (ablation) | MuZero · Ms Pacman (Atari), 50 simulations | episodic return | Helped |
| Use the regularized policy solution as the learning target (ablation) | MuZero · Ms Pacman (Atari), low versus high simulation budgets | episodic return | Mixed |
| Use the regularized policy solution during search (ablation) | MuZero · Ms Pacman (Atari), low and high simulation budgets | episodic return | Helped |
Closest papers by meaning
- Convex Regularization in Monte-Carlo Tree Search (ICML 2021 · similarity 0.92)
- Recursive Monte-Carlo Tree Search (ICML 2026 · similarity 0.91)
- Feedback-Based Tree Search for Reinforcement Learning (ICML 2018 · similarity 0.90)
- Online Robust Reinforcement Learning Through Monte-Carlo Planning (ICML 2025 · similarity 0.88)
- Monte Carlo Tree Search in the Presence of Transition Uncertainty (AAAI 2024 · similarity 0.86)
- SeeA*: Efficient Exploration-Enhanced A* Search by Selective Sampling (NeurIPS 2024 · 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.