Offline Reinforcement Learning on D4RL Gym hopper-medium
104.8Normalized ReturnRORL
Evaluation Results
| Method | Links | |
|---|---|---|
| RORLImplementation Source=Ours, Ensemble Size=102022.06 | 104.8 | |
| SPQRBase algorithm=EDAC2024.01 | 103.8 | |
| EDAC-10Implementation Source=Reproduced, Ensemble Size=102022.06 | 103.6 | |
| SSCQL2025.10 | 102.4 | |
| EDACImplementation=Ours2021.10 | 101.6 | |
| EDACImplementation Source=Paper2022.06 | 101.6 | |
| EDAC2024.01 | 101.6 | |
| SAC-NImplementation=Ours2021.10 | 100.3 | |
| SAC-Min2024.01 | 100.3 | |
| SPQRBase algorithm=SAC-Min2024.01 | 100.2 | |
| QIPO-OT2025.10 | 94 | |
| GTP2025.10 | 90.3 | |
| QIPO-Diff2025.10 | 89.5 | |
| CQLImplementation=Original Paper2021.10 | 86.6 | |
| CQL-Min2024.01 | 86.6 | |
| IPDEvaluation Protocol=Tenfold Episode Evaluation2026.03 | 81.6 | |
| MXQLDomain=Gym Locomotion, Hyperparameter Tuning=individually tuned2025.11 | 80.9 | |
| DDEvaluation Protocol=Tenfold Episode Evaluation2026.03 | 79.3 | |
| QTEvaluation Protocol=Tenfold Episode Evaluation2026.03 | 78 | |
| BRACImplementation=Reproduced2021.10 | 77.8 | |
| QQLDomain=Gym Locomotion, Hyperparameter Tuning=consistent2025.11 | 77.3 | |
| PBRL2022.06 | 75.3 | |
| XQLDomain=Gym Locomotion, Hyperparameter Tuning=individually tuned2025.11 | 74.2 | |
| CQLEvaluation Protocol=Tenfold Episode Evaluation2026.03 | 69.4 | |
| PDiTSeeds=52026.06 | 68.2 | |
| DTEvaluation Protocol=Tenfold Episode Evaluation2026.03 | 67.6 | |
| DRIVESeeds=52026.06 | 67.3 | |
| QDTEvaluation Protocol=Tenfold Episode Evaluation2026.03 | 66.5 | |
| IQLEvaluation Protocol=Tenfold Episode Evaluation2026.03 | 66.3 | |
| IQLDomain=Gym Locomotion, Hyperparameter Tuning=individually tuned2025.11 | 66.3 | |
| IQLSeeds=52026.06 | 66.3 | |
| EDTEvaluation Protocol=Tenfold Episode Evaluation2026.03 | 64.5 | |
| DTSeeds=52026.06 | 63.6 | |
| ReinformerEvaluation Protocol=Tenfold Episode Evaluation2026.03 | 63.5 | |
| CQLImplementation=Reproduced2021.10 | 61.9 | |
| CQL2022.06 | 61.9 | |
| FQL2025.10 | 60.6 | |
| BCSeeds=52026.06 | 59.9 | |
| BCQImplementation=Reproduced2021.10 | 59.4 | |
| TD3+BCDomain=Gym Locomotion, Hyperparameter Tuning=individually tuned2025.11 | 59.3 | |
| TD3+BCSeeds=52026.06 | 59.3 | |
| CQLDomain=Gym Locomotion, Hyperparameter Tuning=individually tuned2025.11 | 58.5 | |
| CQLSeeds=52026.06 | 58 | |
| BEARImplementation=Reproduced2021.10 | 55.3 | |
| BCImplementation=Reproduced2021.10 | 54.1 | |
| BC2022.06 | 54.1 | |
| BC2024.01 | 54.1 | |
| BCDomain=Gym Locomotion, Hyperparameter Tuning=individually tuned2025.11 | 52.9 | |
| MORELImplementation=Reproduced2021.10 | 52.6 | |
| BEARSeeds=52026.06 | 52.1 | |
| UWACImplementation=Reproduced2021.10 | 46.2 | |
| SACImplementation=Reproduced2021.10 | 0.8 | |
| SAC-10Implementation Source=Reproduced, Ensemble Size=102022.06 | 0.8 | |
| REMImplementation=Reproduced2021.10 | 0.7 |