ResearchBenchmarksMulti-agent Reinforcement Learning on GRF CA(hard)Multi-agent Reinforcement Learning on GRF CA(hard)Follow77.38Success RateMAPPO0.253620.276840.360.3232Mar 2, 2021Evaluation resultsMethodMethodLinksSuccess RateMAPPO2021.0377.38TiKickuses pretrained models...uses pretrained models=true2021.0373.09CDS2021.0358.35QMix2021.033.22