Research map / Reinforcement learning / Visual and representation RL
LightZero: A Unified Benchmark for Monte Carlo Tree Search in General Sequential Decision Scenarios
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 |
|---|---|---|---|
| add self-supervised consistency loss (ablation) | MuZero · Atari (six games) | return | Helped |
| add self-supervised consistency loss (ablation) | MuZero · MsPacman | return | No clear effect |
| predict value_prefix instead of reward (EfficientZero) (ablation) | MuZero w/ SSL · Atari | return | Mixed |
| predict value_prefix instead of reward (ablation) | MuZero w/ SSL · MiniGrid KeyCorridorS3R3 sparse reward | return | Hurt |
| use AlphaZero with simulator instead of MuZero | MuZero · Connect4, Gomoku 6x6 | sample efficiency | Helped |
| larger action space dimension for Sampled EfficientZero | Sampled EfficientZero · Breakout (4) vs MsPacman (9) | return | Hurt |
| Gaussian policy representation instead of factored discretization | Sampled EfficientZero · continuous control and MuJoCo | return | Helped |
| Gumbel MuZero instead of MuZero at few simulations | MuZero · Gomoku, LunarLander, Atari | return | Helped |
| Gumbel MuZero instead of MuZero at high simulation counts | MuZero · Gomoku, LunarLander, Atari | return | Hurt |
| Stochastic MuZero instead of MuZero | MuZero · 2048 num_chances=2 | return | Helped |
| increase stochasticity num_chances 2 to 5 | Stochastic MuZero · 2048 | return | Hurt |
| double simulation budget (NaiveDoubleSimulation) (ablation) | MuZero · MiniGrid KeyCorridorS3R3 / FourRooms | return | No clear effect |
Closest papers by meaning
- MALinZero: Efficient Low-Dimensional Search for Mastering Complex Multi-Agent Planning (NeurIPS 2025 · similarity 0.88)
- Threshold UCT: Cost-Constrained Monte Carlo Tree Search with Pareto Curves (AAAI 2025 · similarity 0.88)
- Online Robust Reinforcement Learning Through Monte-Carlo Planning (ICML 2025 · similarity 0.88)
- Multiagent Gumbel MuZero: Efficient Planning in Combinatorial Action Spaces (AAAI 2024 · similarity 0.88)
- Speculative Monte-Carlo Tree Search (NeurIPS 2024 · similarity 0.87)
- Monte Carlo Tree Search in the Presence of Transition Uncertainty (AAAI 2024 · 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.