Multi-Objective Reinforcement Learning on mo-walker2d v5
6,520,000Hypervolume (HV)MORL/D
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| MORL/D2026.03 | 6,520,000 | — | — | 96.9 | |
| MOPDERL2026.03 | 3,470,000 | — | — | 103.2 | |
| MAPEX2026.03 | 3,090,000 | — | — | 157.5 | |
| PRISMReward Sparsity=Extreme (first objective)2026.02 | 47,700 | 120.43 | 59.35 | — | |
| OracleReward Sparsity=None (Dense)2026.02 | 42,100 | 107.58 | 53.22 | — | |
| BaselineReward Sparsity=Extreme (first objective)2026.02 | 33,400 | 82.13 | 39.18 | — |