List your company
ResearchFormerly WizWand

Offline Reinforcement Learning on D4RL walker-medium-replay v2 (test)

82.3Normalized RewardVDT
-2.5192819.5011141.521563.54189Jan 22, 2026
Updated 8mo ago

Evaluation results

MethodLinks
VDT2026.0182.3
TD3+BC2026.0181.8
DDT2026.0177.6
LSDT2026.0174.7
IQL2026.0173.9
DT2026.0166.9
CQL2026.0126.7
BRAC-v2026.010.9
TD3+BC2026.010.837
VDT2026.010.816
BRAC-v2026.010.811
LSDT2026.010.81
CQL2026.010.792
DDT2026.010.786
IQL2026.010.783
DT2026.010.743