List your company

Offline Reinforcement Learning on D4RL MuJoCo v2 (test)

77.6HalfCheetah-Medium ScoreRAMBO
34.4445.64556.8568.055Apr 26, 2022Dec 30, 2022Sep 5, 2023May 11, 2024Jan 15, 2025Sep 21, 2025May 28, 2026

Evaluation results

MethodLinks
RAMBOMethod Category=Model-based, Normalization Procedure=[14], Evaluation Window=last 10 iterations, Seeds=52022.0477.64021.611.592.886.968.996.68593.783.368.3826.2
RAMBO^OFFMethod Category=Model-based, Normalization Procedure=[14], Evaluation Window=last 10 iterations, Seeds=52022.047133.515.50.291.289.16797.688.579.389.563.1785.5
MOPOMethod Category=Model-based, Normalization Procedure=[14], Evaluation Window=last 10 iterations, Seeds=52022.0469.535.44.14.248-0.268.239.169.472.73.3-0.3413.4
SAC-N2026.0568.2———40.887.560.7100.37999101.3114.983.5
EDAC2026.0567.7———101.793.462.199.787.1104.8105.2114.892.9
ReBRAC2026.0564———102.385.851.29584.2103.8109.5111.989.7
COMBOMethod Category=Model-based, Normalization Procedure=[14], Evaluation Window=last 10 iterations, Seeds=52022.0454.238.817.9794.975.555.173.15690111.196.1769.7
RepB-SDEMethod Category=Model-based, Normalization Procedure=[14], Evaluation Window=last 10 iterations, Seeds=52022.0449.132.98.621.13472.157.562.249.855.482.688.8614.1
CQLMethod Category=Model-free, Normalization Procedure=[14], Evaluation Window=last 10 iterations, Seeds=52022.044919.66.72.466.683.847.19788.290.8106.8109.4767.4
UNIQaverages over seeds=0-22026.0548.9———75.685.546101.689.494.8111.8112.985.2
TD3+BCMethod Category=Model-free, Normalization Procedure=[14], Evaluation Window=last 10 iterations, Seeds=52022.0448.3118.51.659.383.744.660.981.890.798110.1698.5
IQL2026.0548.3———67.580.944.597.482.294.7107.4111.781.6
TD3+BC2026.0548.1———60.482.744.864.485.690.8101.211076.4
IQLMethod Category=Model-free, Normalization Procedure=[14], Evaluation Window=last 10 iterations, Seeds=52022.0447.4———66.378.344.294.773.986.791.5109.6692.6
CQL2026.0547———59.180.84595.173.195.699.3109.678.3
BC2026.0542.4———53.563.235.729.821.855.952.39950.4
DT2026.0542.2———65.167.638.981.859.991.6110.4107.173.8
MOReLMethod Category=Model-based, Normalization Procedure=[14], Evaluation Window=last 10 iterations, Seeds=52022.0442.125.653.637.395.477.840.293.649.853.3108.795.6773
BCMethod Category=Model-free, Normalization Procedure=[14], Evaluation Window=last 10 iterations, Seeds=52022.0436.12.19.81.6296.638.411.811.335.8111.96.4300.8