Research map / Multi-agent systems and games / Games and equilibria
Towards Unifying Behavioral and Response Diversity for Open-ended Learning in Zero-sum 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 |
|---|---|---|---|
| add behavioral and response diversity regularizers to P-PSRO (ablation) | P-PSRO · Non-transitive mixture game, PE(5) | population effectivity x100: 40.20 -> 40.54 | Helped |
| add behavioral and response diversity regularizers to P-PSRO (ablation) | P-PSRO · Non-transitive mixture game, PE(25) | population effectivity x100: -12.51 -> -12.18 | Helped |
| add both regularizers vs best single regularizer (ablation) | P-PSRO · PE(15) | population effectivity x100: 11.63 -> 11.57 | No clear effect |
| add behavioral diversity only to P-PSRO (ablation) | P-PSRO · Non-transitive mixture game | exploitability x100: 13.21 -> 41.132 | Hurt |
| add both regularizers to P-PSRO (ablation) | P-PSRO · Non-transitive mixture game | exploitability x100: 13.21 -> 13.26 | No clear effect |
| add behavioral diversity only to P-PSRO (ablation) | P-PSRO · Non-transitive mixture game, PE(5) | population effectivity x100: 40.20 -> 40.19 | No clear effect |
| use too large diversity weight on BD in PSRO (ablation) | PSRO w. BD · AlphaStar game | exploitability x100: 14.64 -> 42.37 | Hurt |
| use too large diversity weight on RD in PSRO (ablation) | PSRO w. RD · AlphaStar game | exploitability x100: 14.77 -> 31.29 | Hurt |
| use too small RD weight (ablation) | PSRO w. RD · AlphaStar game | exploitability x100: 14.06 -> 16.23 | Hurt |
| use PSRO with BD&RD instead of PSRO | PSRO · Google Research Football | win rate vs baseline: 0.78 win rate | Helped |
| use P-PSRO with diversity regularizers instead of PSRO / PSRO_rN | PSRO · Non-transitive mixture game, PE(5) | population effectivity x100: -2.11 -> 40.54 | Helped |
Closest papers by meaning
- Learning Diverse Risk Preferences in Population-Based Self-Play (AAAI 2024 · similarity 0.89)
- Controlling Behavioral Diversity in Multi-Agent Reinforcement Learning (ICML 2024 · similarity 0.88)
- A-PSRO: A Unified Strategy Learning Method with Advantage Metric for Normal-form Games (ICML 2025 · similarity 0.87)
- DGPO: Discovering Multiple Strategies with Diversity-Guided Policy Optimization (AAAI 2024 · similarity 0.87)
- From Average-Iterate to Last-Iterate Convergence in Games: A Reduction and Its Applications (NeurIPS 2025 · similarity 0.86)
- Stability of Multi-Agent Learning in Competitive Networks: Delaying the Onset of Chaos (AAAI 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.