List your company
ResearchFormerly WizWand

Offline Reinforcement Learning on OGBench antmaze-giant-navigate-singletask task1-v0 to task5-v0

52.2ScoreGFP
-2.08812.00626.140.194Dec 3, 2025Dec 26, 2025Jan 19, 2026Feb 12, 2026Mar 7, 2026Mar 31, 2026Apr 24, 2026
Updated 6mo ago

Evaluation results

MethodLinks
GFPTask=task2-v02025.1252.2
ReBRACTask=task5-v02025.1249
DROLPolicy Class=One-step Method, K-selection protocol=benchmark-specific2026.0445
GFPTask=task5-v02025.1243.2
ReBRACTask=task3-v02025.1234
ReBRACTask=task1-v02025.1227
ReBRACPolicy Class=Gaussian Policy2026.0426
IQLTask=task5-v02025.1219
GFPTask=task4-v02025.1217.8
ReBRACTask=task2-v02025.1216
FQLTask=task5-v02025.1216
FQLTask=task4-v02025.1214
GFPTask=task3-v02025.1213.7
IFQLTask=task5-v02025.1213
GFPTask=task1-v02025.1212.6
DeFlowPolicy Class=Multi-step / Iterative Policy2026.0412
FQLTask=task2-v02025.129
FBRACTask=task4-v02025.129
SORLPolicy Class=Multi-step / Iterative Policy2026.049
FQLPolicy Class=One-step Method2026.049
FBRACTask=task5-v02025.126
ReBRACTask=task4-v02025.125
FQLTask=task1-v02025.124
FBRACTask=task2-v02025.124
IQLPolicy Class=Gaussian Policy2026.044
FBRACPolicy Class=Multi-step / Iterative Policy2026.044
IFQLPolicy Class=Multi-step / Iterative Policy2026.043
DROLPolicy Class=One-step Method, K=162026.043
IQLTask=task2-v02025.121
BCTask=task5-v02025.121
BCTask=task1-v02025.120
BCPolicy Class=Gaussian Policy2026.040
FAWACPolicy Class=Multi-step / Iterative Policy2026.040