Research map / Multi-agent systems and games / Cooperative multi-agent RL
Promoting Coordination through Policy Regularization in Multi-Agent Deep Reinforcement Learning
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 CoachReg synchronized sub-policy selection | MADDPG · COMPROMISE | mean return over agents: 18.1 -> 31.1 | Helped |
| add CoachReg synchronized sub-policy selection | MADDPG · BOUNCE | mean return over agents: 4.0 -> 7.4 | Helped |
| add CoachReg synchronized sub-policy selection | MADDPG · SPREAD | mean return over agents: 159 -> 210 | Helped |
| add TeamReg | MADDPG · SPREAD | mean return over agents: 159 -> 216 | Helped |
| add TeamReg | MADDPG · BOUNCE | mean return over agents: 4.0 -> 5.8 | Helped |
| add TeamReg | MADDPG · COMPROMISE | mean return over agents: 18.1 -> 8.8 | Hurt |
| add TeamReg | MADDPG · CHASE | mean return over agents: 834 -> 917 | No clear effect |
| add predictability term (lambda2>0) to agent modelling (ablation) | TeamReg · SPREAD | mean return over agents: 183 -> 216 | Helped |
| add predictability term (lambda2>0) to agent modelling (ablation) | TeamReg · COMPROMISE | mean return over agents: 12.9 -> 8.8 | Hurt |
| add synchronization losses to policy mask (ablation) | CoachReg · COMPROMISE | mean return over agents: 18.4 -> 31.1 | Helped |
| add synchronization losses to policy mask (ablation) | CoachReg · BOUNCE | mean return over agents: 3.7 -> 7.4 | Helped |
| add synchronization losses to policy mask (ablation) | CoachReg · SPREAD | mean return over agents: 221 -> 210 | No clear effect |
Closest papers by meaning
- Multi-Agent Actor-Critic for Mixed Cooperative-Competitive Environments (NeurIPS 2017 · similarity 0.90)
- Celebrating Diversity in Shared Multi-Agent Reinforcement Learning (NeurIPS 2021 · similarity 0.90)
- Mutual-Information Regularized Multi-Agent Policy Iteration (NeurIPS 2023 · similarity 0.89)
- Multi-agent cooperation through learning-aware policy gradients (ICLR 2025 · similarity 0.88)
- Graph Diffusion for Robust Multi-Agent Coordination (ICML 2025 · similarity 0.88)
- Backpropagation Through Agents (AAAI 2024 · similarity 0.88)
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.