Goal-conditioned Reinforcement Learning (G2_balanced) on hopper medium v2
1.564Goal-conditioned ReturnPrior Only
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Prior Only2026.04 | 1.564 | 5.931 | |
| KL-Reg2026.04 | 1.553 | 2.081 | |
| PoE2026.04 | 1.553 | 2.09 | |
| Frozen2026.04 | 1.541 | 0 | |
| Additive2026.04 | 1.537 | 1.502 |