Offline Reinforcement Learning on Hopper Medium Noise 0
95Normalized ReturnCQL
Evaluation Results
| Method | Links | |
|---|---|---|
| CQLModel Category=Model-Free2026.02 | 95 | |
| IQLModel Category=Model-Free2026.02 | 94.7 | |
| L-MAPModel Category=Model-Based2026.02 | 90.8 | |
| 1R2RModel Category=Model-Based2026.02 | 89.67 | |
| I-TAPModel Category=Model-Based2026.02 | 86.57 | |
| I-TAPModel Category=Model-Based2026.02 | 81.94 | |
| TAPModel Category=Model-Based2026.02 | 80.92 | |
| TAPModel Category=Model-Based2026.02 | 69.14 | |
| IQLModel Category=Model-Free2026.02 | 66.3 | |
| L-MAPModel Category=Model-Based2026.02 | 61.65 | |
| DTModel Category=Model-Free2026.02 | 58.14 | |
| CQLModel Category=Model-Free2026.02 | 58 | |
| 1R2RModel Category=Model-Based2026.02 | 55.49 | |
| DTModel Category=Model-Free2026.02 | 43.01 |