List your company

Offline Reinforcement Learning on D4RL walker2d-medium-expert

116.1Normalized ScorePSPO
104.348107.399110.45113.501Jun 16, 2021Apr 9, 2022Feb 1, 2023Nov 26, 2023Sep 19, 2024Jul 14, 2025May 8, 2026

Evaluation results

MethodLinks
PSPOAveraged across=4 random seeds2026.05116.1
RRPINumber of random seeds=42026.03115.7
AMGNumber of random seeds=42026.03114.9
ADM2026.05114.9
EDACMethod Type=Model-free2023.10114.7
DMGNumber of random seeds=42026.03114.4
DMG2026.05114.4
ATACSeeds=52023.10114.2
ATACMethod Type=Model-free2023.10114.2
Exp. WeightAlgorithmic Template=One-step2021.06112.9
Count-MORLMethod Type=Model-based2023.10112.3
EPQNumber of random seeds=42026.03112
EPQ2026.05112
PMDBNumber of random seeds=42026.03111.9
PMDBMethod Type=Model-based2023.10111.9
PMDB2026.05111.9
Easy BCQAlgorithmic Template=One-step2021.06111.8
IQLAccess=True reward functions2024.04111.7
IDQL2025.12111.6
DMambasource=Lv et al., 20242023.06111.6
CROPMethod Type=Model-based2023.10111.6
Repb-SDE2021.10111.2
Fu et al.Algorithmic Template=Iterative2021.06111
Rev. KL Reg.Algorithmic Template=One-step2021.06110.9
ROMI-BCQ2021.10110.7
BCQ2021.10110.7
MPDiffuserranking=false2025.12110.7
MPDiffuser+Rankranking=true2025.12110.7
LSTMOracle status=false, Implementation source=Kim et al. [28]2023.05110.4
PTAccess=Scripted preferences2024.04110.4
Planner2025.12110.4
CACNFE=22026.03110.4
CBCNFE=22026.03110.4
DMMModel Category=SSM-based2026.02110.2
PTOracle status=false, Implementation source=Kim et al. [28]2023.05110.1
TD3+BCModel Category=Value-based2026.02110.1
DQLNFE=52026.03110.1
ROMI-CQL2021.10109.7
IQLSeeds=52023.10109.6
IQL2023.06109.6
IQL2023.10109.6
IQL2026.03109.6
IQLMethod Type=Model-free2023.10109.6
IQL2025.09109.6
CQL2021.10109.4
DCModel Category=SSM-based2026.02109.2
MROracle status=false, Implementation source=reimpl.2023.05109.1
CQLNumber of random seeds=42026.03109.1
CQL2026.05109.1
DTModel Category=Transformer-based2026.02109
CQL2023.06108.8
DDImplementation=Original2023.06108.8
CQL2023.10108.8
DDCategory=Diffusion Planner2024.05108.8
Decision Diffuser2025.12108.8
CQLModel Category=Value-based2026.02108.8
CQL2023.06108.8
MambaDM2023.06108.8
CQL2026.03108.8
CQL2025.09108.8
Decision Diffuser2025.09108.8
RATE2023.06108.7
CABI+TD3-BCMethod Type=Model-based2023.10108.6
WFDiffuser2025.09108.6
IPLOracle status=false, Implementation source=Ours2023.05108.51
KDPNFE=12026.03108.5
Diffuser2023.06108.4
Diffuser2023.10108.4
Diffuser2025.12108.4
MOReL2025.09108.4
MROracle status=false, Implementation source=Kim et al. [28]2023.05108.3
DMModel Category=SSM-based2026.02108.3
DMambasource=Ota, 20242023.06108.3
ProposedSeeds=52023.10108.2
AdaptDiffuserCategory=Diffusion Planner2024.05108.2
DT2023.06108.1
DT2023.10108.1
DT2025.12108.1
DT2023.06108.1
DT2026.03108.1
Decision Stacks2023.06108
RGGplanning seeds=152023.10107.8
IQLOracle status=true, Implementation source=Kim et al. [28]2023.05107.8
EDTModel Category=Transformer-based2026.02107.8
RGG+planning seeds=152023.10107.7
BC2023.06107.5
BC2023.10107.5
BC2025.12107.5
MOReLNumber of random seeds=42026.03107.5
BC2026.03107.5
MOReL2026.05107.5
BC2025.09107.5
TAP2023.06107.4
DiffuserCategory=Diffusion Planner2024.05106.9
DiffuserNFE=202026.03106.9
DS4Model Category=SSM-based2026.02105.7
OPRLAccess=Scripted preferences2024.04105.4
DDImplementation=reproduced2023.06105.2
OPPOAccess=Scripted preferences, Implementation=Reported2024.04105
CAMPAccess=Scripted preferences2024.04104.8