List your company
ResearchFormerly WizWand

Offline-to-online Reinforcement Learning on D4RL antmaze-medium-play

81.7OSRCQL
-3.26818.79140.8562.909May 11, 2026
Updated 5mo ago

Evaluation results

MethodLinks
CQL2026.0581.798
RankQvariant=+SAC2026.0581.798.7
CQLvariant=+SAC2026.0578.398.3
Cal-QLvariant=+SAC2026.057598.9
RankQ2026.057597.7
Cal-QL2026.0571.797.8
Hybrid RL2026.0513.398.3
SACvariant=+OFF2026.0511.798.8
SAC2026.0500