AI Paper Atlas

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

Towards Complete Multi-Agent Coordination Policy Learning via Denoising Maximum Entropy Optimization

Guanghao Li, Lei Yuan, Ruiqi Xue, Hengchang Zhang, Jianhong Wang, Yi-Chen Li, Yang Yu

ICML 2026 · Multi-agent systems and games › Cooperative multi-agent RL · approach: Cooperative MARL algorithms

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 changedWhereMeasuredEffect
add complete division (CD) of the observation space to the diffusion shared policy (ablation)Dspic · HalfCheetah 6x1average convergence performanceImproved
add complete division (CD) of the observation space to the diffusion shared policy (ablation)Dspic · Walker2d 6x1stability of trainingImproved
add the diffusion policy on top of CD (ablation)Dspic · HalfCheetah 6x1 and Walker2d 6x1stability and asymptotic performanceImproved
use complete division instead of concatenating agent ID to a shared policy (ablation)Dspic without CD · HalfCheetah 6x1 and Walker2d 6x1average performance and stabilityImproved
use complete division instead of concatenating role embedding z_i to the observation (ablation)Dspic without CD · HalfCheetah 6x1 and Walker2d 6x1average performance and stabilityImproved
add Conservative Temperature Adaptation (CTA) (ablation)Dspic · SMAC 3s5z vs 3s6ztraining stability / convergenceImproved
use Dspic with CD instead of a shared policy without CDshared policy · LBF case studysolution length (steps): 8 -> 5Improved
use maximum entropy objective (HASAC) instead of HAPPOheterogeneous-agent baseline · LBFperformanceHurt

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