W-REINFORCE
| Method | Links | |||||
|---|---|---|---|---|---|---|
2026.05 | 33.7 | 98.3 | 37.2 | 99.7 | 67.2 | |
2026.05 | 33.5 | 97.6 | 37.1 | 99.5 | 66.9 | |
2026.05 | 33 | 97.3 | 36 | 99.5 | 66.4 | |
2026.05 | 32.7 | 97.1 | 36.1 | 99 | 66.2 | |
2026.05 | 32.2 | 95.8 | 36.6 | 99.8 | 66.1 | |
2026.05 | 31.2 | 97.7 | 34.4 | 98.7 | 65.5 | |
2026.05 | 30.2 | 97 | 33.5 | 98.9 | 64.9 | |
2026.05 | 30.1 | 97 | 32.6 | 99.4 | 64.8 | |
2026.05 | 29.9 | 93.3 | 32.4 | 98.3 | 63.5 | |
2026.05 | 29.3 | 94 | 32.7 | 97.9 | 63.5 | |
2026.05 | 27.4 | 93.1 | 28.8 | 97.8 | 61.8 | |
2026.05 | 26.3 | 90.3 | 34.6 | 99.6 | 62.7 | |
2026.05 | 23.2 | 92.2 | 26.4 | 97.3 | 59.8 |