atlas

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

SMACv2: An Improved Benchmark for Cooperative Multi-Agent Reinforcement Learning

NeurIPS 2023 · Multi-agent systems and games › Cooperative multi-agent RL · approach: MARL benchmarks and grouping · 28 citations (OpenAlex)

Read the paperCheck your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's title only (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
Remove random start positions (ablation)MAPPO on SMACv2 5-unit scenarios · SMACv2win rateHelped
Use only surround or reflect starting-position scenarios rather than combining them (ablation)MAPPO on SMACv2 5-unit scenarios · SMACv2win rateNo clear effect
Restrict unit composition to a single non-special unit type (ablation)MAPPO on SMACv2 5-unit scenarios · SMACv2win rateHelped
Use fixed unit sight and attack ranges instead of varied SMACv2 ranges (ablation)MAPPO on SMACv2 5-unit scenarios · SMACv2win rateHelped
Replace the recurrent network with a feed-forward network (ablation)MAPPO on SMACv2 5-unit scenarios · SMACv2win rateHurt

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