Goal-conditioned Reinforcement Learning (G3_efficient) on hopper medium v2
31.8Goal-conditioned ReturnPrior Only
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Prior Only2026.04 | 31.8 | 4.524 | |
| PoE2026.04 | 31.5 | 2.099 | |
| KL-Reg2026.04 | 31.4 | 2.143 | |
| Additive2026.04 | 31 | 1.123 | |
| Frozen2026.04 | 30.9 | 0 |