List your company
ResearchFormerly WizWand

Offline Reinforcement Learning on D4RL antmaze-umaze (diverse)

93.5Normalized ScoreFAC
-4.57220.88946.3571.811Jun 12, 2021Apr 9, 2022Feb 5, 2023Dec 4, 2023Sep 30, 2024Jul 29, 2025May 27, 2026
Updated 5mo ago

Evaluation results

MethodLinks
FACPolicy Type=Flow, Seeds=82026.0293.5
Decision Mamba2024.0690
FQLPolicy Type=Flow, Seeds=82026.0289
GCIQL2023.0788.8
ReBRACPolicy Type=Gaussian, Seeds=82026.0288.3
CQLPolicy Type=Gaussian, Seeds=82026.0284
SRPOPolicy Type=Diffusion, Seeds=82026.0282.1
GTP2025.1081.9
QQLDomain=AntMaze, Hyperparameter Tuning=consistent2025.1181.3
IDQLPolicy Type=Diffusion, Seeds=82026.0280.2
MCQPolicy Type=Gaussian, Seeds=82026.0280
IQLDomain=AntMaze, Hyperparameter Tuning=individually tuned2025.1179.5
EPQPolicy Type=Gaussian, Seeds=82026.0278.3
CACPolicy Type=Diffusion, Seeds=82026.0277.6
XQLDomain=AntMaze, Hyperparameter Tuning=individually tuned2025.1177.2
SPAR-PROJLearning Paradigm=Ours2026.0576.7
QIPO-OTPolicy Type=Diffusion, Seeds=82026.0276.1
QIPO-OT2025.1076.1
DWSL2023.0774.6
QIPO-Diff2025.1073.9
TD3+BCEvaluations=final 10, Seeds=5, State feature normalization=false2021.0671.4
TD3+BCPolicy Type=Gaussian, Seeds=82026.0271.4
TD3+BCDomain=AntMaze, Hyperparameter Tuning=individually tuned2025.1171.3
IQL2023.0771.2
GCPC2023.0771.2
LaMo2024.0670
RvS-G2023.0766.2
SAC-RNDPolicy Type=Gaussian, Seeds=82026.0266
BCDomain=AntMaze, Hyperparameter Tuning=individually tuned2025.1164.8
ROMI-CQL2021.1063.6
BC2023.0763.4
SPAR-MLPLearning Paradigm=Ours2026.0563.3
IQLPolicy Type=Gaussian, Seeds=82026.0262.2
PLASLearning Paradigm=Policy Gradient Guidance2026.0562
IDQLLearning Paradigm=In-Support Learning2026.0562
ROMI-BCQ2021.1061.2
BC2024.0661
DT2024.0661
COMBO2021.1057.3
Diff-QLLearning Paradigm=Policy Gradient Guidance2026.0557.3
WGCSL2023.0755.6
CQL-AWLearning Paradigm=In-Support Learning2026.0554
CQLDomain=AntMaze, Hyperparameter Tuning=individually tuned2025.1153.8
DT2023.0753.4
MXQLDomain=AntMaze, Hyperparameter Tuning=individually tuned2025.1153.2
BAIL2021.1052
EQLLearning Paradigm=In-Support Learning2026.0550.8
BiTrajDiffAlgorithm=CQL, Augmentation Method=Ours2025.0648.8
BCQ2021.1048
BiTrajDiffAlgorithm=DT, Augmentation Method=Ours2025.0647.8
CQL2023.0747.4
BC2021.1047
RTDiffAlgorithm=DT, Augmentation Method=RTDiff2025.0646.4
SyntherAlgorithm=DT, Augmentation Method=Synther2025.0644.8
DiffStitchAlgorithm=DT, Augmentation Method=DiffStitch2025.0642.6
SyntherAlgorithm=CQL, Augmentation Method=Synther2025.0642.4
BaseAlgorithm=DT, Augmentation Method=Base2025.0642
RTDiffAlgorithm=CQL, Augmentation Method=RTDiff2025.0640.6
BaseLearning Paradigm=Ours2026.0540
DiffStitchAlgorithm=CQL, Augmentation Method=DiffStitch2025.0639.2
CQLLearning Paradigm=Policy Gradient Guidance2026.0539.2
BEAR2021.1037.7
IQLLearning Paradigm=In-Support Learning2026.0537.3
BaseAlgorithm=CQL, Augmentation Method=Base2025.0631.6
BCQLearning Paradigm=Policy Gradient Guidance2026.0528
LAPOLearning Paradigm=In-Support Learning2026.0528
CQL2021.106.3
TD3+BCLearning Paradigm=Policy Gradient Guidance2026.051.7
BRAC-P2021.100
BRAC-v2021.100
MOPO2021.100
MOREL2021.100
Repb-SDE2021.100
AWACLearning Paradigm=In-Support Learning2026.05-0.8