Multi-objective Reinforcement Learning on MO-PointMaze 2obj
6.67ESRAETDICE
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| AETDICEObjective optimized=ESR2026.06 | 6.67 | 11.9 | 8.73 | 164.86 | |
| ESR-IQL2026.06 | 6.56 | 11.59 | 8.56 | 153.78 | |
| AETDICEObjective optimized=BSR0.52026.06 | 5.91 | 11.93 | 9.01 | 211.79 | |
| AETDICEObjective optimized=SER2026.06 | 4.06 | 11.82 | 9.74 | 284.05 | |
| Linear ScalarizationScalarization weights (w)=[0.75, 0.25]2026.06 | 3.67 | 8.91 | 4.53 | 415.88 | |
| Linear ScalarizationScalarization weights (w)=[0.5, 0.5]2026.06 | 3.64 | 8.77 | 3.81 | 417.01 | |
| Linear ScalarizationScalarization weights (w)=[1.0, 0.0]2026.06 | 3.63 | 8.71 | 3.82 | 409.62 |