Cross-Domain Offline Reinforcement Learning on MuJoCo Hopper (mr/me to m/me/e v2)
76.5Normalized ScoreV2A
Evaluation Results
| Method | Links | |
|---|---|---|
| V2ABase Alignment=IGDF, Source=hopper-me, Target=me2026.05 | 76.5 | |
| V2ABase Alignment=OTDF, Source=hopper-me, Target=me2026.05 | 74.3 | |
| V2ABase Alignment=OTDF, Source=hopper-me, Target=e2026.05 | 72.4 | |
| V2ABase Alignment=OTDF, Source=hopper-me, Target=m2026.05 | 65.7 | |
| V2ABase Alignment=IGDF, Source=hopper-me, Target=e2026.05 | 64.8 | |
| V2ABase Alignment=OTDF, Source=hopper-mr, Target=e2026.05 | 64.6 | |
| V2ABase Alignment=IGDF, Source=hopper-mr, Target=e2026.05 | 61.7 | |
| V2ABase Alignment=IGDF, Source=hopper-mr, Target=m2026.05 | 58.8 | |
| V2ABase Alignment=OTDF, Source=hopper-mr, Target=m2026.05 | 58.3 | |
| V2ABase Alignment=IGDF, Source=hopper-mr, Target=me2026.05 | 56.8 | |
| V2ABase Alignment=IGDF, Source=hopper-me, Target=m2026.05 | 56.4 | |
| V2ABase Alignment=OTDF, Source=hopper-mr, Target=me2026.05 | 52.7 |