Research map / Reinforcement learning / Policy gradient and actor-critic
Convex Regularization in Monte-Carlo Tree Search
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 |
|---|---|---|---|
| Use Tsallis entropy regularization instead of relative entropy regularization | MCTS on Asteroids · Asteroids | average score: 1,486.40 -> 1,642.10 | Helped |
| Use Tsallis entropy regularization instead of maximum entropy regularization | MCTS on Phoenix · Phoenix | average score: 3,098.30 -> 3,975.30 | Helped |
| Use Tsallis entropy regularization instead of UCT | MCTS on Phoenix · Phoenix | average score: 2,747.30 -> 3,975.30 | Helped |
| Use Tsallis entropy regularization instead of MaxMCTS | MCTS on Phoenix · Phoenix | average score: 2,626.60 -> 3,975.30 | Helped |
| Replace non-regularized PUCT with convex-regularized MCTS | AlphaGo on Atari · Atari 22 games | games with highest mean: 6/22 -> 22/22 | Helped |
| Use sparse Tsallis-entropy policies rather than maximum- or relative-entropy policies | MCTS on the synthetic tree · Synthetic tree | root value estimation convergence | Helped |
| Use Tsallis entropy rather than maximum or relative entropy | MCTS on the synthetic tree · Synthetic tree | regret | Helped |
| Use regularized E3W selection rather than UCT | MCTS on the synthetic tree · Synthetic tree | regret | Hurt |
Closest papers by meaning
- Monte-Carlo Tree Search as Regularized Policy Optimization (ICML 2020 · similarity 0.92)
- Maximum Entropy Monte-Carlo Planning (NeurIPS 2019 · similarity 0.90)
- Online Robust Reinforcement Learning Through Monte-Carlo Planning (ICML 2025 · similarity 0.89)
- Monte Carlo Tree Search in the Presence of Transition Uncertainty (AAAI 2024 · similarity 0.87)
- Convergence Theorems for Entropy-Regularized and Distributional Reinforcement Learning (NeurIPS 2025 · similarity 0.87)
- Monte-Carlo Tree Search with Uncertainty Propagation via Optimal Transport (ICML 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.