Offline Meta Reinforcement Learning on Ant-dir (out-of-distribution)
410.7Average ReturnSPC
Evaluation Results
| Method | Links | |
|---|---|---|
| SPCZero-shot=true2026.03 | 410.7 | |
| UNICORN-SSZero-shot=true2026.03 | 405.9 | |
| CSROZero-shot=true2026.03 | 399.2 | |
| FOCALZero-shot=true2026.03 | 368.8 | |
| UNICORN-SUPZero-shot=true2026.03 | 211.4 | |
| DORAZero-shot=true2026.03 | 156.8 |