Goal-conditioned Reinforcement Learning (G3_efficient) on hopper medium-expert v2
0.336Goal-conditioned ReturnFrozen
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Frozen2026.04 | 0.336 | 0 | |
| Additive2026.04 | 0.335 | 1.676 | |
| KL-Reg2026.04 | 0.334 | 2.18 | |
| PoE2026.04 | 0.333 | 1.262 | |
| Prior Only2026.04 | 0.331 | 6.676 |