AI Paper Atlas

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

SPD: Synergy Pattern Diversifying Oriented Unsupervised Multi-agent Reinforcement Learning

Yuhang Jiang, Jianzhun Shao, Shuncheng He, Hongchang Zhang, Xiangyang Ji

NeurIPS 2022 · Multi-agent systems and games › Cooperative multi-agent RL · approach: Intrinsic-reward exploration for MARL · 1 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 SPD synergy-pattern pseudo-reward instead of DIAYN/WURL state-diversity rewardsQMIX unsupervised MARL · MPE SimpleTagNormalized proximal discrepancy of synergy patterns: (4.0 ± 2.7) × 10−2 -> 1.0 ± 0.1Improved
Use SPD instead of DIAYN/WURL on state identifiabilityQMIX unsupervised MARL · MPE SimpleTagDiscriminator success rate (%): 59.9 ± 0.9 -> 58.6 ± 0.7No clear effect
Initialize downstream QMIX with SPD-pretrained model instead of training QMIX from scratchQMIX · Google Research Football 3_vs_1 with keeper, four downstream taskswinning rateImproved
Initialize QMIX with WURL or DIAYN pretrained modelQMIX · GRF downstream taskswinning rateNo clear effect
Scale pseudo-reward factor beta_r over 0.1, 1.0, 10.0 (default 10.0) (ablation)SPD · MPE SimpleTagproximal discrepancy of synergy patternsMixed
Decompose episode return into per-time-step pseudo-rewards (vs sparse final-step return) (ablation)SPD · MPE SimpleTagproximal discrepancy of synergy patternsImproved
Add optimal-transport matching rho* instead of rho(t)=t (ablation)SPD · MPE SimpleTagproximal discrepancy of synergy patternsImproved
Increase Kuhn-Munkres iterations from 25 to 100 (ablation)SPD · MPE SimpleTagproximal discrepancy of synergy patterns: 25 -> 100Improved

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