Research map / Multi-agent systems and games / Cooperative multi-agent RL
Consensus Learning for Cooperative Multi-Agent 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 inferred discrete consensus to each agent's input | MADDPG · MPE: Cooperative Predator-Prey, Cooperative Navigation, Cooperative Pantomime | mean episode return | Helped |
| Add inferred discrete consensus to each agent's input | QMIX · SMAC: 5m_vs_6m and 27m_vs_30m | win ratio | Helped |
| Add inferred discrete consensus to each agent's input | QMIX · SMAC: easy scenarios | win ratio | Helped |
| Add inferred discrete consensus to each agent's input | QMIX · Google Research Football: academy_corner and other evaluated scenarios | performance | Helped |
| Add inferred discrete consensus to each agent's input | VDN · SMAC: MMM2 without global state | performance | Helped |
| Add inferred discrete consensus to each agent's input | VDN · SMAC: 2c_vs_64zg | performance | No clear effect |
| Use historical hidden states instead of raw local observations as consensus-builder input (ablation) | COLA-QMIX · SMAC | performance | Helped |
| Set the consensus category count to K=32 instead of smaller values (ablation) | COLA-QMIX · SMAC: 3s5z | performance | Hurt |
| Use K=32 or K=16 instead of small K (ablation) | COLA-QMIX · SMAC: MMM2 | performance | Helped |
| Use COLA consensus when agents do not need to reach consensus on the global state (ablation) | COLA · SMAC: 2c_vs_64zg | performance | No clear effect |
Closest papers by meaning
- Efficient Communication in Multi-Agent Reinforcement Learning with Implicit Consensus Generation (AAAI 2025 · similarity 0.90)
- Intrinsic Action Tendency Consistency for Cooperative Multi-Agent Reinforcement Learning (AAAI 2024 · similarity 0.88)
- Multi-agent cooperation through learning-aware policy gradients (ICLR 2025 · similarity 0.88)
- Settling Decentralized Multi-Agent Coordinated Exploration by Novelty Sharing (AAAI 2024 · similarity 0.87)
- Multi-Agent Reinforcement Learning with Communication-Constrained Priors (NeurIPS 2025 · similarity 0.87)
- Learning Joint Behaviors with Large Variations (AAAI 2025 · similarity 0.87)
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.