Goal-conditioned Reinforcement Learning (G3_efficient) on walker2d medium v2
0.374Goal-conditioned ReturnPrior Only
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Prior Only2026.04 | 0.374 | 9.688 | |
| KL-Reg2026.04 | 0.368 | 5.367 | |
| PoE2026.04 | 0.368 | 5.286 | |
| Additive2026.04 | 0.358 | 2.427 | |
| Frozen2026.04 | 0.349 | 0 |