List your company

Multi-task Reinforcement Learning on Meta-World MT50 v1 (Fixed)

60Success RateSoft Modularization (Deep)
21.41631.43341.4551.467Mar 30, 2020

Evaluation results

MethodLinks
Soft Modularization (Deep)Network Depth=Deep2020.0360
Soft Modularization (Shallow)Network Depth=Shallow2020.0359.5
Mix-ExpertImplementation=Internal implementation baseline2020.0336.1
MT-MH-SACImplementation=Reported in [43]2020.0335.9
MT-MH-SACImplementation=Internal implementation baseline2020.0335.5
MT-SACImplementation=Internal implementation baseline2020.0331.4
MT-SACImplementation=Reported in [43]2020.0328.8
Hard RoutingImplementation=Internal implementation baseline2020.0322.9