List your company
ResearchFormerly WizWand

Offline Reinforcement Learning on D4RL MuJoCo walker2d-medium-expert

114.2Normalized ScoreMCQ
-4.56826.26657.187.934Feb 20, 2026Mar 10, 2026Mar 28, 2026Apr 15, 2026May 3, 2026May 21, 2026Jun 9, 2026
Updated 4mo ago

Evaluation results

MethodLinks
MCQPolicy Type=Gaussian, Seeds=82026.02114.2
SRPOPolicy Type=Diffusion, Seeds=82026.02114
SRPOPolicy Type=One-Step Flow/Diffusion Policy2026.06114
SPAR-PROJLearning Paradigm=Ours2026.05113.4
BFQPolicy Type=One-Step Flow/Diffusion Policy2026.06113.4
OFQLPolicy Type=One-Step Flow/Diffusion Policy2026.06113
IDQLPolicy Type=Diffusion, Seeds=82026.02112.7
ISEP-FMFlow Matching=true2026.05112.5
EPQPolicy Type=Gaussian, Seeds=82026.02112
SPOTPolicy Type=Gaussian, Seeds=82026.02112
ReBRACPolicy Type=Gaussian, Seeds=82026.02111.6
DQL2026.05111.6
IDQLLearning Paradigm=In-Support Learning2026.05111.6
DQLPolicy Type=Diffusion Policy2026.06111.6
IDQL2026.05111.2
QIPO-OTPolicy Type=Diffusion, Seeds=82026.02110.9
Diff-QLLearning Paradigm=Policy Gradient Guidance2026.05110.9
LAPOLearning Paradigm=In-Support Learning2026.05110.9
QIPO2026.05110.8
QGPO2026.05110.7
IDQLPolicy Type=Diffusion Policy2026.06110.6
CACPolicy Type=Diffusion, Seeds=82026.02110.4
TD3-BCPolicy Type=Gaussian Policy2026.06110.4
EDPPolicy Type=Diffusion Policy2026.06110.4
ISEPFlow Matching=false2026.05110.2
EQLLearning Paradigm=In-Support Learning2026.05110.2
TD3+BCPolicy Type=Gaussian, Seeds=82026.02110.1
TD3+BCLearning Paradigm=Policy Gradient Guidance2026.05110.1
BaseLearning Paradigm=Ours2026.05109.7
IQLPolicy Type=Gaussian, Seeds=82026.02109.6
IQL2026.05109.6
IQLLearning Paradigm=In-Support Learning2026.05109.6
SORLPolicy Type=One-Step Flow/Diffusion Policy2026.06109.1
BCQLearning Paradigm=Policy Gradient Guidance2026.05109
CQL-AWLearning Paradigm=In-Support Learning2026.05109
CQLPolicy Type=Gaussian, Seeds=82026.02108.8
CQLLearning Paradigm=Policy Gradient Guidance2026.05108.8
IQLPolicy Type=Gaussian Policy2026.06108.7
FACPolicy Type=Flow, Seeds=82026.02108.4
BC2026.05107.5
FQLPolicy Type=Flow, Seeds=82026.02106.1
BCPolicy Type=Gaussian Policy2026.06105.4
FQLPolicy Type=One-Step Flow/Diffusion Policy2026.06100.5
CQL2026.0598.7
PLASLearning Paradigm=Policy Gradient Guidance2026.0597.2
AWACLearning Paradigm=In-Support Learning2026.0574.5
SPAR-MLPLearning Paradigm=Ours2026.050