Offline Reinforcement Learning on halfcheetah jump
4,016.6Average ReturnMOPO
Evaluation Results
| Method | Links | |
|---|---|---|
| MOPO2021.02 | 4,016.6 | |
| MOREL2021.02 | 3,228.7 | |
| Batch Max2021.02 | 1,808.6 | |
| CQL2021.02 | 741.1 | |
| Batch Mean2021.02 | -1,022.6 |
Loading the SOTA2 catalog…
| Method | Links | |
|---|---|---|
| MOPO2021.02 | 4,016.6 | |
| MOREL2021.02 | 3,228.7 | |
| Batch Max2021.02 | 1,808.6 | |
| CQL2021.02 | 741.1 | |
| Batch Mean2021.02 | -1,022.6 |