List your company
ResearchFormerly WizWand

Multi-Agent Reinforcement Learning on SMAC 3m v1

100.9Normalized Win RateMAST
8.54832.52456.580.476Sep 28, 2024
Updated 8mo ago

Evaluation results

MethodLinks
MASTAlg.=Q-MIX, Sparsity=95%, Total Size=0.066x, FLOPs (Train)=0.051x, FLOPs (Test)=0.050x2024.09100.9
MASTAlg.=RES, Sparsity=95%, Total Size=0.066x, FLOPs (Train)=0.055x, FLOPs (Test)=0.050x2024.0999.8
MASTAlg.=WQ-MIX, Sparsity=90%, Total Size=0.108x, FLOPs (Train)=0.100x, FLOPs (Test)=0.100x2024.0998.6
TinyAlg.=Q-MIX, Sparsity=95%, Total Size=0.066x, FLOPs (Train)=0.051x, FLOPs (Test)=0.050x2024.0998.3
TinyAlg.=WQ-MIX, Sparsity=90%, Total Size=0.108x, FLOPs (Train)=0.100x, FLOPs (Test)=0.100x2024.0998.3
RLx2Alg.=WQ-MIX, Sparsity=90%, Total Size=0.108x, FLOPs (Train)=0.100x, FLOPs (Test)=0.100x2024.0998
RLx2Alg.=RES, Sparsity=95%, Total Size=0.066x, FLOPs (Train)=0.055x, FLOPs (Test)=0.050x2024.0997.9
SETAlg.=WQ-MIX, Sparsity=90%, Total Size=0.108x, FLOPs (Train)=0.100x, FLOPs (Test)=0.100x2024.0997.8
RigLAlg.=WQ-MIX, Sparsity=90%, Total Size=0.108x, FLOPs (Train)=0.100x, FLOPs (Test)=0.100x2024.0997.8
TinyAlg.=RES, Sparsity=95%, Total Size=0.066x, FLOPs (Train)=0.055x, FLOPs (Test)=0.050x2024.0997.8
SETAlg.=RES, Sparsity=95%, Total Size=0.066x, FLOPs (Train)=0.055x, FLOPs (Test)=0.050x2024.0997.3
SSAlg.=WQ-MIX, Sparsity=90%, Total Size=0.108x, FLOPs (Train)=0.100x, FLOPs (Test)=0.100x2024.0996.9
SETAlg.=Q-MIX, Sparsity=95%, Total Size=0.066x, FLOPs (Train)=0.051x, FLOPs (Test)=0.050x2024.0996
SSAlg.=RES, Sparsity=95%, Total Size=0.066x, FLOPs (Train)=0.055x, FLOPs (Test)=0.050x2024.0995.6
RigLAlg.=Q-MIX, Sparsity=95%, Total Size=0.066x, FLOPs (Train)=0.051x, FLOPs (Test)=0.050x2024.0995.3
SSAlg.=Q-MIX, Sparsity=95%, Total Size=0.066x, FLOPs (Train)=0.051x, FLOPs (Test)=0.050x2024.0991.6
RigLAlg.=RES, Sparsity=95%, Total Size=0.066x, FLOPs (Train)=0.055x, FLOPs (Test)=0.050x2024.0991.1
RLx2Alg.=Q-MIX, Sparsity=95%, Total Size=0.066x, FLOPs (Train)=0.051x, FLOPs (Test)=0.050x2024.0912.1