List your company

Offline Reinforcement Learning on Expert Backward

67.24Final ReturnIQL + SEL
41.895248.475155.05561.6349Feb 20, 2026

Evaluation results

MethodLinks
IQL + SELAlgorithm=IQL + SEL, Number of seeds=52026.0267.24
BC + EGAlgorithm=BC + EG, Number of seeds=52026.0267.24
IQL + EGAlgorithm=IQL + EG, Number of seeds=52026.0267.24
BCAlgorithm=BC, Number of seeds=52026.0267.17
IQLAlgorithm=IQL, Number of seeds=52026.0267.1
IQL + PCGradAlgorithm=IQL + PCGrad, Number of seeds=52026.0267.05
IQL + EGAlgorithm=IQL + EG, Number of seeds=52026.0257.58
IQL + PCGradAlgorithm=IQL + PCGrad, Number of seeds=52026.0255.94
IQL + SELAlgorithm=IQL + SEL, Number of seeds=52026.0255.73
TD3+BC + EGAlgorithm=TD3+BC + EG, Number of seeds=52026.0255.67
TD3+BCAlgorithm=TD3+BC, Number of seeds=52026.0252.31
TD3+BC + EGAlgorithm=TD3+BC + EG, Number of seeds=52026.0251.98
IQLAlgorithm=IQL, Number of seeds=52026.0251.86
BC + EGAlgorithm=BC + EG, Number of seeds=52026.0248.64
TD3+BCAlgorithm=TD3+BC, Number of seeds=52026.0247.94
BCAlgorithm=BC, Number of seeds=52026.0242.87