Research map / Learning theory and algorithms / Submodular and game algorithms
Efficient Subgame Refinement for Extensive-form Games
Zhenxing Ge, Zheng Xu, Tianyu Ding, Wenbin Li, Yang Gao
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 |
|---|---|---|---|
| increase generated block size k in diversity-based GS2 (ablation) | GS2-D · Leduc, Liar's Dice, Goofspiel (k=3..5) | exploitability | Improved |
| increase generated states k in GS2 (k=10,20,30) (ablation) | GS2 on GuanDan · GuanDan vs second champion | average score | Improved |
| use diversity-based instead of random generation function (ablation) | GS2 · 10-card and 12-card Goofspiel | exploitability | Improved |
| apply GS2 with diversity generation instead of time-limited 1-KLSS | Blueprint refinement · 10-card and 12-card Goofspiel | exploitability | Improved |
| apply GS2-D instead of unlimited-time 1-KLSS | Blueprint refinement · 5-card Goofspiel and Liar's Dice | exploitability | No clear effect |
| apply unsafe subgame solving | Blueprint · 3-rank limit Leduc poker | exploitability | Improved |
| apply unsafe subgame solving | Blueprint · Liar's Dice | exploitability | Hurt |
| apply GS2 refinement to a coarse blueprint under 3-second limit | Blueprint · research games | exploitability | Improved |
| apply GS2 refinement (k=10,20,30) to DMC blueprint | GuanDan agent · GuanDan vs second champion | average score | Improved |
| apply GS2 refinement to the blueprint | GuanDan agent · GuanDan vs first champion | average score | Improved |
Closest papers by meaning
- Refining Subgames in Large Imperfect Information Games (AAAI 2016 · similarity 0.90)
- GAVEL: Generating Games via Evolution and Language Models (NeurIPS 2024 · similarity 0.86)
- Mastering Board Games by External and Internal Planning with Language Models (ICML 2025 · similarity 0.85)
- GTBench: Uncovering the Strategic Reasoning Capabilities of LLMs via Game-Theoretic Evaluations (NeurIPS 2024 · similarity 0.85)
- The Update-Equivalence Framework for Decision-Time Planning (ICLR 2024 · similarity 0.84)
- Solving Hierarchical Information-Sharing Dec-POMDPs: An Extensive-Form Game Approach (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.
