AI Paper Atlas

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

Heterogeneous Skill Learning for Multi-agent Tasks

Yuntao Liu, Yuan Li, Xinhai Xu, Yong Dou, Donghong Liu

NeurIPS 2022 · Multi-agent systems and games › Cooperative multi-agent RL · approach: MARL benchmarks and grouping

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 HSL (skill representation, skill selector, skill-based policy learning) instead of CDSHSL · SMAC 6h_vs_8ztest win rate: around 10%Improved
Use HSL instead of CDSHSL · SMAC 5m_vs_6mtest win rate: around 20%Improved
Use HSL instead of CDS on a large-scale map (many agents)HSL · SMAC 27m_vs_30mtest win rateImproved
Use HSL instead of RODE in heterogeneous-agent maps (MMM2, 3s5z_vs_3s6z)HSL · SMAC MMM2 and 3s5z_vs_3s6ztest win rateImproved
Use HSL instead of HSD (skill discovery from local observation only)HSL · SMAC Super Hard mapstest win rateImproved
Use HSL instead of QMIX/QPLEX/role-based methodsHSL · GRF 3 vs 1 with keeper, Hard Counter-attack, Cornertest win rateImproved
Use HSL instead of CDS with many agentsHSL · GRF Cornertest win rateImproved
Use HSL instead of role-based methods and baselinesHSL · GoBigger 3 vs 3 and 3 vs 3 with thorntest win rateImproved
Add the skill selector instead of random skill assignment (ablation)HSL · SMAC 6h_vs_8z, GRF Hard Counter-attack, GoBigger 3 vs 3test win rateImproved
Add the learned skill representation instead of raw states for skill assignment (ablation)HSL · SMAC 6h_vs_8z, GRF Hard Counter-attack, GoBigger 3 vs 3test win rateImproved
Add the intrinsic (entropy) reward to the skill selector objective (ablation)HSL · GRF Hard Counter-attack and GoBigger 3 vs 3test win rateImproved
Add the intrinsic reward to the skill selector (ablation)HSL · SMAC 6h_vs_8ztest win rateNo clear effect

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