List your company

Offline Reinforcement Learning on walker2d medium-replay

101.3Normalized ScoreQ-ALIGN DT
-4.05223.29950.6578.001Jun 16, 2021Apr 13, 2022Feb 8, 2023Dec 6, 2023Oct 2, 2024Jul 30, 2025May 27, 2026

Evaluation results

MethodLinks
Q-ALIGN DT2026.05101.3
ProposedSeeds=52023.1099.6
QT2026.0598.5
Anti-exploration Method2026.0296.3
QCS2026.0594.1
DMG2026.0289.7
SAC-RND2026.0288.7
ATACSeeds=52023.1086.5
GPC-SAC2026.0286.2
DM2026.0585.5
O-DICE2026.0283.6
CQL2026.0281.8
TD3+BC2026.0581.8
C-LAPaction-conditioned data percentage=100%, variant=Oracle2025.1281.2
DT2026.0579.4
CGDT2026.0578.1
DC2026.0576.6
LAWMaction-conditioned data percentage=5%2025.1275.9
RADT2026.0575.9
C-LAPaction-conditioned data percentage=5%2025.1275.7
I-TAPNoise=3.52026.0275.11
LSDT2026.0574.7
IQLSeeds=52023.1073.9
IQL2026.0273.8
IQL2026.0573.7
L-MAPNoise=3.52026.0271.98
I-TAPNoise=9.52026.0271.9
Rev. KL Reg.Algorithmic Template=One-step2021.0671.6
Easy BCQAlgorithmic Template=One-step2021.0666.4
L-MAPNoise=9.52026.0264.68
TD3+BCaction-conditioned data percentage=100%, variant=Oracle2025.1263.9
I-TAPNoise=152026.0260.7
DTReward Specification=Step-wise2024.0260.2
DTNoise=3.52026.0258.49
QDTReward Specification=Step-wise2024.0258.2
COMBOSeeds=52023.1056
IDM-TD3+BCaction-conditioned data percentage=5%2025.1252.3
DTReward Specification=Final Return2024.0251.6
L-MAPNoise=152026.0250.64
Exp. WeightAlgorithmic Template=One-step2021.0649.3
DTNoise=9.52026.0242.78
TD3+BCaction-conditioned data percentage=5%2025.1235.9
TD3+BCSeeds=52023.1034.4
DTNoise=152026.0229.91
QDTReward Specification=Final Return2024.0229.6
CQLReward Specification=Step-wise2024.0226.7
Fu et al.Algorithmic Template=Iterative2021.0626.7
BCAlgorithmic Template=One-step2021.0623.9
OptiDICESeeds=52023.1021.6
CQLSeeds=52023.1020.8
BEARSeeds=52023.1019.2
TD3-CVAE2026.0215.4
BCQSeeds=52023.1015
DVDF-OTDF2025.124.8
IGDF2025.124.4
OTDF2025.124.1
IQL2025.123.6
CQLReward Specification=Final Return2024.023.2
DVDF-IGDF2025.123
DARA2025.122.9
BOSA2025.120