Offline Reinforcement Learning on AntMaze Medium Diverse v0
85.4Normalized ScoreQDQ
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| QDQrandom seeds=52024.10 | 85.4 | — | — | |
| IQLrandom seeds=52024.10 | 70 | — | — | |
| CQLrandom seeds=52024.10 | 53.7 | — | — | |
| TD3+BCrandom seeds=52024.10 | 3 | — | — | |
| AWACrandom seeds=52024.10 | 0.7 | — | — | |
| BCrandom seeds=52024.10 | 0 | — | — | |
| DTrandom seeds=52024.10 | 0 | — | — | |
| Onestep RLrandom seeds=52024.10 | 0 | — | — | |
| diff-QLseeds=102023.10 | — | 6.4 | 82 | |
| IQLseeds=10, checkpoint_selection=last 10 average2023.10 | — | 70 | — | |
| MSGseeds=10, checkpoint_selection=last 10 average2023.10 | — | 83 | — | |
| SEEMseeds=102023.10 | — | 83.9 | 90.7 | |
| sfBCseeds=10, checkpoint_selection=last 10 average2023.10 | — | 90 | — | |
| TD3+BCseeds=10, checkpoint_selection=last 10 average2023.10 | — | 0 | — |