List your company
ResearchFormerly WizWand

Multi-Agent Reinforcement Learning on SMAC 6h* v1

104.9Normalized Win RateMAST
6.131.7557.483.05Sep 28, 2024
Updated 8mo ago

Evaluation results

MethodLinks
MASTAlg.=RES, Sparsity=85%, Total Size=0.155x, FLOPs (Train)=0.151x, FLOPs (Test)=0.150x2024.09104.9
MASTAlg.=WQ-MIX, Sparsity=90%, Total Size=0.104x, FLOPs (Train)=0.100x, FLOPs (Test)=0.100x2024.0998.4
MASTAlg.=Q-MIX, Sparsity=90%, Total Size=0.106x, FLOPs (Train)=0.100x, FLOPs (Test)=0.100x2024.0997.6
TinyAlg.=RES, Sparsity=85%, Total Size=0.155x, FLOPs (Train)=0.151x, FLOPs (Test)=0.150x2024.0983.3
RLx2Alg.=RES, Sparsity=85%, Total Size=0.155x, FLOPs (Train)=0.151x, FLOPs (Test)=0.150x2024.0972.7
SETAlg.=Q-MIX, Sparsity=90%, Total Size=0.106x, FLOPs (Train)=0.100x, FLOPs (Test)=0.100x2024.0967.1
TinyAlg.=Q-MIX, Sparsity=90%, Total Size=0.106x, FLOPs (Train)=0.100x, FLOPs (Test)=0.100x2024.0958.2
RLx2Alg.=WQ-MIX, Sparsity=90%, Total Size=0.104x, FLOPs (Train)=0.100x, FLOPs (Test)=0.100x2024.0952.8
TinyAlg.=WQ-MIX, Sparsity=90%, Total Size=0.104x, FLOPs (Train)=0.100x, FLOPs (Test)=0.100x2024.0951
RigLAlg.=Q-MIX, Sparsity=90%, Total Size=0.106x, FLOPs (Train)=0.100x, FLOPs (Test)=0.100x2024.0948.7
SETAlg.=WQ-MIX, Sparsity=90%, Total Size=0.104x, FLOPs (Train)=0.100x, FLOPs (Test)=0.100x2024.0944.1
SETAlg.=RES, Sparsity=85%, Total Size=0.155x, FLOPs (Train)=0.151x, FLOPs (Test)=0.150x2024.0944.1
RigLAlg.=WQ-MIX, Sparsity=90%, Total Size=0.104x, FLOPs (Train)=0.100x, FLOPs (Test)=0.100x2024.0941
SSAlg.=Q-MIX, Sparsity=90%, Total Size=0.106x, FLOPs (Train)=0.100x, FLOPs (Test)=0.100x2024.0940.2
SSAlg.=RES, Sparsity=85%, Total Size=0.155x, FLOPs (Train)=0.151x, FLOPs (Test)=0.150x2024.0939.1
RigLAlg.=RES, Sparsity=85%, Total Size=0.155x, FLOPs (Train)=0.151x, FLOPs (Test)=0.150x2024.0935.3
SSAlg.=WQ-MIX, Sparsity=90%, Total Size=0.104x, FLOPs (Train)=0.100x, FLOPs (Test)=0.100x2024.0929.6
RLx2Alg.=Q-MIX, Sparsity=90%, Total Size=0.106x, FLOPs (Train)=0.100x, FLOPs (Test)=0.100x2024.099.9