Offline Reinforcement Learning on walker2d medium-replay
101.3Normalized ScoreQ-ALIGN DT
Evaluation Results
| Method | Links | |
|---|---|---|
| Q-ALIGN DT2026.05 | 101.3 | |
| ProposedSeeds=52023.10 | 99.6 | |
| QT2026.05 | 98.5 | |
| Anti-exploration Method2026.02 | 96.3 | |
| QCS2026.05 | 94.1 | |
| DMG2026.02 | 89.7 | |
| SAC-RND2026.02 | 88.7 | |
| ATACSeeds=52023.10 | 86.5 | |
| GPC-SAC2026.02 | 86.2 | |
| DM2026.05 | 85.5 | |
| O-DICE2026.02 | 83.6 | |
| CQL2026.02 | 81.8 | |
| TD3+BC2026.05 | 81.8 | |
| C-LAPaction-conditioned data percentage=100%, variant=Oracle2025.12 | 81.2 | |
| DT2026.05 | 79.4 | |
| CGDT2026.05 | 78.1 | |
| DC2026.05 | 76.6 | |
| LAWMaction-conditioned data percentage=5%2025.12 | 75.9 | |
| RADT2026.05 | 75.9 | |
| C-LAPaction-conditioned data percentage=5%2025.12 | 75.7 | |
| I-TAPNoise=3.52026.02 | 75.11 | |
| LSDT2026.05 | 74.7 | |
| IQLSeeds=52023.10 | 73.9 | |
| IQL2026.02 | 73.8 | |
| IQL2026.05 | 73.7 | |
| L-MAPNoise=3.52026.02 | 71.98 | |
| I-TAPNoise=9.52026.02 | 71.9 | |
| Rev. KL Reg.Algorithmic Template=One-step2021.06 | 71.6 | |
| Easy BCQAlgorithmic Template=One-step2021.06 | 66.4 | |
| L-MAPNoise=9.52026.02 | 64.68 | |
| TD3+BCaction-conditioned data percentage=100%, variant=Oracle2025.12 | 63.9 | |
| I-TAPNoise=152026.02 | 60.7 | |
| DTReward Specification=Step-wise2024.02 | 60.2 | |
| DTNoise=3.52026.02 | 58.49 | |
| QDTReward Specification=Step-wise2024.02 | 58.2 | |
| COMBOSeeds=52023.10 | 56 | |
| IDM-TD3+BCaction-conditioned data percentage=5%2025.12 | 52.3 | |
| DTReward Specification=Final Return2024.02 | 51.6 | |
| L-MAPNoise=152026.02 | 50.64 | |
| Exp. WeightAlgorithmic Template=One-step2021.06 | 49.3 | |
| DTNoise=9.52026.02 | 42.78 | |
| TD3+BCaction-conditioned data percentage=5%2025.12 | 35.9 | |
| TD3+BCSeeds=52023.10 | 34.4 | |
| DTNoise=152026.02 | 29.91 | |
| QDTReward Specification=Final Return2024.02 | 29.6 | |
| CQLReward Specification=Step-wise2024.02 | 26.7 | |
| Fu et al.Algorithmic Template=Iterative2021.06 | 26.7 | |
| BCAlgorithmic Template=One-step2021.06 | 23.9 | |
| OptiDICESeeds=52023.10 | 21.6 | |
| CQLSeeds=52023.10 | 20.8 | |
| BEARSeeds=52023.10 | 19.2 | |
| TD3-CVAE2026.02 | 15.4 | |
| BCQSeeds=52023.10 | 15 | |
| DVDF-OTDF2025.12 | 4.8 | |
| IGDF2025.12 | 4.4 | |
| OTDF2025.12 | 4.1 | |
| IQL2025.12 | 3.6 | |
| CQLReward Specification=Final Return2024.02 | 3.2 | |
| DVDF-IGDF2025.12 | 3 | |
| DARA2025.12 | 2.9 | |
| BOSA2025.12 | 0 |