Research map / Multi-agent systems and games / Games and equilibria
Policy Space Diversity for Non-Transitive Games
Jian Yao, Weiming Liu, Haobo Fu, Yaodong Yang, Stephen Marcus McAleer, QIANG FU, Yang Wei
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 |
|---|---|---|---|
| Use PSD-PSRO instead of PSRO | PSRO · Goofspiel (8 point cards) | win rate: 0.422$\pm$0.025 -> 0.578$\pm$0.025 | Improved |
| Use PSD-PSRO instead of PSRO$_{rN}$ | PSRO$_{rN}$ · Goofspiel (8 point cards) | win rate: 0.358$\pm$0.019 -> 0.642$\pm$0.019 | Improved |
| Use PSD-PSRO instead of P-PSRO | P-PSRO · Goofspiel (8 point cards) | win rate: 0.370$\pm$0.031 -> 0.630$\pm$0.031 | Improved |
| Add diversity regularization with λ=2 rather than λ=0 (ablation) | PSD-PSRO · Non-Transitive Mixture Game | exploitability × 100: 40.88 $\pm$ 2.18 -> 1.72 $\pm$ 0.54 | Improved |
| Set diversity weight to λ=50 rather than λ=2 (ablation) | PSD-PSRO · Non-Transitive Mixture Game | exploitability × 100: 1.72 $\pm$ 0.54 -> 8.22 $\pm$ 1.73 | Hurt |
| Add diversity regularization with λ=0.5 rather than λ=0 (ablation) | PSD-PSRO · Leduc Poker | exploitability: 0.590 $\pm$ 0.017 -> 0.356 $\pm$ 0.012 | Improved |
| Set diversity weight to λ=5 rather than λ=0.5 (ablation) | PSD-PSRO · Leduc Poker | exploitability: 0.356 $\pm$ 0.012 -> 0.664 $\pm$ 0.062 | Hurt |
Closest papers by meaning
- Global Policy-Space Response Oracles for Two-Player Zero-Sum Games (ICML 2026 · similarity 0.90)
- A-PSRO: A Unified Strategy Learning Method with Advantage Metric for Normal-form Games (ICML 2025 · similarity 0.88)
- Learning Diverse Risk Preferences in Population-Based Self-Play (AAAI 2024 · similarity 0.86)
- Optimistic Policy Gradient in Multi-Player Markov Games with a Single Controller: Convergence beyond the Minty Property (AAAI 2024 · similarity 0.85)
- Controlling Behavioral Diversity in Multi-Agent Reinforcement Learning (ICML 2024 · similarity 0.85)
- A Policy-Gradient Approach to Solving Imperfect-Information Games with Best-Iterate Convergence (ICLR 2025 · similarity 0.85)
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.
