AI Paper Atlas

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

Offline Multi-Agent Reinforcement Learning via Sequential Score Decomposition

Dan Qiao, Wenhao Li, Shanchao Yang, Hongyuan Zha, Baoxiang Wang

ICML 2026 · Multi-agent systems and games › Cooperative multi-agent RL · approach: Offline and networked MARL · 0 citations (OpenAlex)

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
replace independent per-agent behavior regularization with sequential score decomposition (ablation)behavior-regularized policy optimization · Bandit toy (2-Gaussian mixture)evaluation reward: 0±1 -> 1±0Improved
use sequential conditional scores rather than centralized critic with marginal regularization (ablation)BRPO-CTDE · Bandit toy (2-Gaussian mixture)evaluation reward: 0±1 -> 1±0Improved
sequential score decomposition versus joint action learning (ablation)OMSD · Bandit toy (2-Gaussian mixture)evaluation reward: 1±0 vs 1±0No clear effect
OMSD versus best baselines on MPE, averaged over 9 datasetsoffline MARL on MPE · MPE (9 datasets, average)average normalized score: 87.3 -> 126.7Improved
OMSD on lower-quality (random, medium) MPE dataoffline MARL on MPE · MPE Predator Prey Randomnormalized score: 78.5 -> 133.9Improved
OMSD on expert Cooperative Navigation dataoffline MARL on MPE · MPE Cooperative Navigation Expertnormalized score: 126.3 -> 102.3Hurt
OMSD on MaMuJoCo 2-Ant and 4-Ant Good datasetsoffline MARL on MaMuJoCo · MaMuJoCo 2-Ant Goodepisode return: 2946 -> 2714Hurt
OMSD on OMIGA MaMuJoCo benchmarkoffline MARL on MaMuJoCo · MaMuJoCo OMIGA datasets (average)average episode return: 1954.74 -> 3400Improved
add OMSD post-training to pretrained IQL critic/policypretrained IQL · MaMuJoCo (average across datasets)episode returnImproved
randomize agent update order in sequential decomposition (ablation)OMSD · OMIGA Hopper-v2episode returnNo clear effect
vary regularization coefficient beta over 0.001 to 0.5 (ablation)OMSD · MaMuJoCo / MPEepisode returnMixed

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