AI Paper Atlas

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

An Efficient Transfer Learning Framework for Multiagent Reinforcement Learning

Tianpei Yang, Weixun Wang, Hongyao Tang, Jianye HAO, Zhaopeng Meng, Hangyu Mao, Dong Li, Wulong Liu, Yingfeng Chen, Yujing Hu, Changjie Fan, Chengwei Zhang

NeurIPS 2021 · Multi-agent systems and games › Cooperative multi-agent RL · approach: Cooperative MARL algorithms · 6 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
Use MAPTF knowledge transfer instead of PPO or DVMPPO ghosts · Pac-Man OpenClassicaverage discounted reward: +2.5 -> +3Improved
Use MAPTF instead of DVM policy distillation as scenario difficulty increasesPPO · Pac-Man MediumClassic and OriginalClassicaverage rewardImproved
Add MAPTF to PPO as number of agents growsPPO · MPE predator-preyaverage rewardImproved
Add MAPTF to MADDPG and QMIXMADDPG, QMIX · MPE tasksaverage rewardImproved
Replace SRO with traditional single-agent option learning (ablation)MAPTF · MPE (MADDPG, QMIX)average rewardImproved
Use MAPTF w/o SRO instead of training from scratch (ablation)MADDPG, QMIX · MPEaverage rewardImproved
Add parameter sharing to MAPTF-PPO (ablation)MAPTF with PPO · MPE predator-preyaverage rewardMixed
Compare MAPTF without PS against PPO with PS (ablation)MAPTF · MPE predator-preyaverage rewardImproved

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