List your company
ResearchFormerly WizWand

Offline Reinforcement Learning on Walker2d medium-expert

121.4Normalized ScoreQ-ALIGN DT
5.75235.77665.895.824Dec 2, 2025Dec 31, 2025Jan 29, 2026Feb 28, 2026Mar 29, 2026Apr 27, 2026May 27, 2026
Updated 5mo ago

Evaluation results

MethodLinks
Q-ALIGN DT2026.05121.4
QCS2026.05116.6
DMG2026.02114.4
Anti-exploration Method2026.02112.8
DM2026.05112.7
QT2026.05112.6
GPC-SAC2026.02111.7
O-DICE2026.02110.8
TD3+BC2026.05110.1
LSDT2026.05109.8
RADT2026.05109.7
IQL2026.02109.6
IQL2026.05109.6
DC2026.05109.6
CGDT2026.05109.3
DT2026.05108.1
CQL2026.02107.9
SAC-RND2026.02105
I-TAPNoise=3.52026.0298.8
L-MAPNoise=3.52026.0296.53
I-TAPNoise=9.52026.0294.29
L-MAPNoise=9.52026.0292.31
I-TAPNoise=152026.0288.26
L-MAPNoise=152026.0287.21
TD3-CVAE2026.0284.9
DROCO2025.1278.9
IGDF2025.1275.7
OTDF2025.1266.7
DTNoise=3.52026.0265.07
DTNoise=9.52026.0256.23
BOSA2025.1246.7
DTNoise=152026.0244.07
DARA2025.1242.2
IQL*variant=tuned/reproduced2025.1227.6
DVDF-OTDF2025.1223
CQL*variant=tuned/reproduced2025.1222.4
DVDF-IGDF2025.1220.9
IQL2025.1216.7
OTDF2025.1215.4
BOSA2025.1212.8
IGDF2025.1212.6
DARA2025.1210.2