List your company
ResearchFormerly WizWand

Offline Reinforcement Learning on D4RL antmaze-med (diverse)

86.7Normalized ScoreEPQ
-3.46819.94143.3566.759Oct 1, 2021Jun 25, 2022Mar 19, 2023Dec 11, 2023Sep 3, 2024May 28, 2025Feb 20, 2026
Updated 8mo ago

Evaluation results

MethodLinks
EPQPolicy Type=Gaussian, Seeds=82026.0286.7
QIPO-OTPolicy Type=Diffusion, Seeds=82026.0286.4
FACPolicy Type=Flow, Seeds=82026.0285
IDQLPolicy Type=Diffusion, Seeds=82026.0284.8
ReBRACPolicy Type=Gaussian, Seeds=82026.0276.3
SRPOPolicy Type=Diffusion, Seeds=82026.0275
SAC-RNDPolicy Type=Gaussian, Seeds=82026.0274.7
FQLPolicy Type=Flow, Seeds=82026.0271
IQLPolicy Type=Gaussian, Seeds=82026.0270
CQL2021.1061.5
CQLPolicy Type=Gaussian, Seeds=82026.0253.7
MCQPolicy Type=Gaussian, Seeds=82026.0237.5
ROMI-BCQ2021.1027.3
ROMI-CQL2021.1020.9
TD3+BCPolicy Type=Gaussian, Seeds=82026.023
BAIL2021.102
BC2021.100
BCQ2021.100
BEAR2021.100
BRAC-P2021.100
BRAC-v2021.100
MOPO2021.100
MOREL2021.100
Repb-SDE2021.100
COMBO2021.100
CACPolicy Type=Diffusion, Seeds=82026.020