Research map / Multi-agent systems and games / Games and equilibria
Split Moves for 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 |
|---|---|---|---|
| Split moves with Mod instead of using orthodox moves | Flat Monte-Carlo · Breakthrough | nodal states/sec.: 2,388,827 (100%) -> 5,717,972 (239%) | Helped |
| Use ModShift rather than Mod splitting (ablation) | Flat Monte-Carlo · Breakthrough | nodal states/sec.: 5,717,972 (239%) -> 7,558,711 (316%) | Helped |
| Split moves with Mod instead of using orthodox moves | Flat Monte-Carlo · Chess | nodal states/sec.: 308,157 (100%) -> 1,556,036 (505%) | Helped |
| Use ModPlus rather than Mod splitting (ablation) | Flat Monte-Carlo · Chess | nodal states/sec.: 1,556,036 (505%) -> 1,390,428 (451%) | Hurt |
| Split moves with Mod instead of using orthodox moves | Flat Monte-Carlo · Chess | simulations/sec.: 976 -> 5,971 | Helped |
| Use ModPlus rather than Mod splitting (ablation) | Flat Monte-Carlo · Chess | simulations/sec.: 5,971 -> 5,341 | Hurt |
| Use semisplit moves in simulations (ablation) | MCTS · 12 board games | playing strength | Helped |
| Use semisplit moves in the MCTS tree (ablation) | MCTS · 12 board games | playing strength | Mixed |
| Increase split granularity from Mod to ModPlus (ablation) | Semisplit MCTS · 12 board games | playing strength | Helped |
| Increase split granularity from ModPlus to ModShift (ablation) | Semisplit MCTS · 12 board games | playing strength | Hurt |
| Use the nodal rather than raw expansion variant (ablation) | Semisplit MCTS · 12 board games | playing strength | No clear effect |
| Use roll-up rather than ordinary semisplit tree handling (ablation) | Semisplit MCTS · 12 board games | playing strength | Mixed |
Closest papers by meaning
- Nested Monte Carlo Search for Two-Player Games (AAAI 2016 · similarity 0.88)
- LightZero: A Unified Benchmark for Monte Carlo Tree Search in General Sequential Decision Scenarios (NeurIPS 2023 · similarity 0.86)
- Mastering Board Games by External and Internal Planning with Language Models (ICML 2025 · similarity 0.85)
- Implicit Search via Discrete Diffusion: A Study on Chess (ICLR 2025 · similarity 0.84)
- MALinZero: Efficient Low-Dimensional Search for Mastering Complex Multi-Agent Planning (NeurIPS 2025 · similarity 0.84)
- Mastering Zero-Shot Interactions in Cooperative and Competitive Simultaneous Games (ICML 2024 · 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.