AI Paper Atlas

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

Learning Nearly Decomposable Value Functions Via Communication Minimization

Tonghan Wang*, Jianhao Wang*, Chongyi Zheng, Chongjie Zhang

ICLR 2020 · Multi-agent systems and games › Cooperative multi-agent RL · approach: Value factorization · 24 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
Cut all messages (ablation)NDQ · Independent searchaverage team reward per episode: 96.0 -> 96.0No clear effect
Cut all messages (ablation)TarMAC + QMIX · Independent searchaverage team reward per episode: 96.0 -> 96.0No clear effect
Cut 80% of messages using learned message distributions (ablation)NDQ · StarCraft II micromanagementperformanceNo clear effect
Cut 80% of messages by removing the smallest attention weights (ablation)QMIX + TarMAC · StarCraft II micromanagementperformanceHurt
Cut all messages (ablation)NDQ · StarCraft II micromanagementwin rateHurt
Increase the communication penalty to β=1.0 (ablation)NDQ · SensorcommunicationMixed
Decrease the communication penalty to β=10^-5 (ablation)NDQ · SensorcommunicationHurt
Use β=10^-3 for message regularization (ablation)NDQ · Sensorcommunication: 80%Improved
Add learned communication to value function factorizationNDQ · StarCraft II micromanagementperformanceImproved
Add information-theoretic message constraints rather than attention-only communicationNDQ vs QMIX + TarMAC · StarCraft II micromanagementperformanceImproved
Add communication to QMIX using TarMACQMIX + TarMAC · StarCraft II micromanagementperformanceHurt
Add minimal communication learned by NDQNDQ · Hallwaywin rate: 100%Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 2 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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