Research map / Learning theory and algorithms / Submodular and game algorithms
Subgame Solving in Adversarial Team Games
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 |
|---|---|---|---|
| apply subgame refinement (alpha=5) to the blueprint | Team-maxmargin · Kuhn K45 | team value: -0.354 -> -0.046 | Helped |
| apply subgame refinement (alpha=5) to the blueprint | Team-maxmargin · Kuhn K38 | team value: -0.152 -> -0.031 | Helped |
| apply subgame refinement (alpha=5) to the blueprint | Team-maxmargin · Leduc L3523 | team value: -6.671 -> -1.151 | Helped |
| apply subgame refinement (alpha=5) to the blueprint | Team-maxmargin · Leduc L3133 | team value: -0.616 -> 0.038 | Helped |
| apply subgame refinement (alpha=5) to the blueprint | Team-maxmargin · Liar's Dice D33 | team value: 0.238 -> 0.279 | Helped |
| apply subgame refinement (alpha=5) to the blueprint | Team-maxmargin · Tricks T3100 | team value: 0.495 -> 0.514 | Helped |
| apply subgame refinement (alpha=5) to the blueprint | Team-maxmargin · Leduc L3151 | team value: -0.624 -> -0.399 | Helped |
| apply subgame refinement (alpha=5) to the blueprint | Team-maxmargin · all test instances | gap reduction: about 65% | Helped |
| limit refinement time to one column generation iteration (alpha=1) | Team-maxmargin · all test instances | gap reduction: about 39% | Helped |
| raise refinement budget from alpha=1 to alpha=5 | Team-maxmargin · all test instances | gap reduction: 39% -> 65% | Helped |
| refinement versus the exact equilibrium | Team-maxmargin · all test instances | team value | No clear effect |
| increase refinement time (alpha) | Team-maxmargin · test suite curves | team value | Mixed |
Closest papers by meaning
- Computational Results for Extensive-Form Adversarial Team Games (AAAI 2018 · similarity 0.92)
- DAG-Based Column Generation for Adversarial Team Games (ICML 2024 · similarity 0.91)
- The Complexity of Two-Team Polymatrix Games with Independent Adversaries (ICLR 2025 · similarity 0.87)
- Learning Equilibria in Adversarial Team Markov Games: A Nonconvex-Hidden-Concave Min-Max Optimization Problem (NeurIPS 2024 · similarity 0.86)
- Opponent Modeling based on Subgoal Inference (NeurIPS 2024 · similarity 0.86)
- Generative Adversarial Equilibrium Solvers (ICLR 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.