atlas

Research map / Multi-agent systems and games / Cooperative multi-agent RL

Multi-Agent Actor-Critic for Mixed Cooperative-Competitive Environments

NeurIPS 2017 · Multi-agent systems and games › Cooperative multi-agent RL · approach: Decentralized actor-critic MARL · 1,002 citations (OpenAlex)

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 changedWhereMeasuredEffect
centralized critic that conditions on all agents' actions (MADDPG) instead of decentralized RLmulti-agent actor-critic · Cooperative communicationlistener reaches targetHelped
approximating other agents' policies online instead of using true policies (ablation)MADDPG · Cooperative communicationaverage rewardNo clear effect
training agents with policy ensembles (K sub-policies) (ablation)MADDPG · Keep-away, cooperative navigation, predator-preywin rate vs single-policy adversariesHelped
centralized critic (MADDPG) vs decentralized DDPGcooperative navigationcollisions per episode (N=2): roughly half the collisionsHelped

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy