Cross-Domain Offline Reinforcement Learning on MuJoCo Ant (ant-mr/ant-me to m/me/e v2)
126.2Normalized ScoreV2A
Evaluation Results
| Method | Links | |
|---|---|---|
| V2ABase Alignment=OTDF, Source=ant-me, Target=e2026.05 | 126.2 | |
| V2ABase Alignment=IGDF, Source=ant-me, Target=e2026.05 | 124.3 | |
| V2ABase Alignment=OTDF, Source=ant-me, Target=me2026.05 | 117.3 | |
| V2ABase Alignment=IGDF, Source=ant-me, Target=me2026.05 | 114 | |
| V2ABase Alignment=IGDF, Source=ant-me, Target=m2026.05 | 96.3 | |
| V2ABase Alignment=OTDF, Source=ant-me, Target=m2026.05 | 93.3 | |
| V2ABase Alignment=OTDF, Source=ant-mr, Target=me2026.05 | 85.6 | |
| V2ABase Alignment=IGDF, Source=ant-mr, Target=me2026.05 | 81.7 | |
| V2ABase Alignment=OTDF, Source=ant-mr, Target=e2026.05 | 78.7 | |
| V2ABase Alignment=IGDF, Source=ant-mr, Target=e2026.05 | 70.2 | |
| V2ABase Alignment=OTDF, Source=ant-mr, Target=m2026.05 | 56.6 | |
| V2ABase Alignment=IGDF, Source=ant-mr, Target=m2026.05 | 54.3 |