incumbent
done · peak unseen_mismatch 96.9
α schedule
α 3.0 ×19 α 3.0 ×19 α 3.0 ×19 α 3.0 ×19 α 3.0 ×19 α 3.0 ×19 α 3.0 ×19 α 3.0 ×19 α 1.0 ×19 α 1.0 ×19 α 1.0 ×19 α 1.0 ×19 α 1.0 ×19 α 1.0 ×19 α 1.0 ×19 α 1.0 ×19 α 2.1429 ×19 α 2.1429 ×19 α 2.1429 ×19 α 2.1429 ×19 α 2.1429 ×19 α 2.1429 ×19 α 2.1429 ×19 α 2.1429 ×19 α 1.0 ×19 α 1.0 ×19 α 1.0 ×19 α 1.0 ×19 α 1.0 ×19 α 1.0 ×19 α 1.0 ×19 α 1.0 ×19 α 1.2857 ×19 α 1.2857 ×19 α 1.2857 ×19 α 1.2857 ×19 α 1.2857 ×18 α 1.2857 ×18 α 1.2857 ×18 α 1.2857 ×18 α 1.0 ×18 α 1.0 ×18 α 1.0 ×18 α 1.0 ×18 α 1.0 ×18 α 1.0 ×18 α 1.0 ×18 α 1.0 ×18
Figures
Results
accuracy by training iteration, mean ± sd over seeds · schedule ends at α = 1.0
seen_* probed at reference α = 1.4 · unseen_* = held-out pairings, uniform (α-independent) · 1000 pairs per condition
| iteration | seen_match | seen_mismatch | unseen_match | unseen_mismatch |
|---|---|---|---|---|
| 25 | 51.4 ± 1.5 | 49.5 ± 1.6 | 49.8 ± 1.9 | 49.7 ± 2.1 |
| 50 | 50.7 ± 1.9 | 49.6 ± 1.4 | 50.1 ± 1.5 | 50.8 ± 1.1 |
| 75 | 51.9 ± 1.8 | 50.1 ± 1.9 | 50.5 ± 1.7 | 50.0 ± 1.9 |
| 100 | 53.8 ± 2.9 | 52.1 ± 2.9 | 51.5 ± 1.8 | 50.0 ± 1.5 |
| 125 | 65.8 ± 10.5 | 60.5 ± 8.1 | 59.9 ± 9.2 | 54.9 ± 5.3 |
| 150 | 79.3 ± 12.4 | 73.6 ± 12.8 | 72.2 ± 14.1 | 66.3 ± 12.7 |
| 175 | 92.1 ± 12.9 | 83.0 ± 15.8 | 89.8 ± 14.6 | 79.1 ± 16.2 |
| 200 | 97.0 ± 11.2 | 80.7 ± 22.7 | 97.0 ± 10.6 | 78.2 ± 21.7 |
| 225 | 97.3 ± 11.4 | 85.9 ± 18.9 | 97.3 ± 11.0 | 84.2 ± 18.6 |
| 250 | 97.4 ± 11.1 | 86.7 ± 22.2 | 97.4 ± 10.4 | 85.2 ± 22.3 |
| 275 | 97.5 ± 10.6 | 88.5 ± 21.4 | 97.4 ± 10.8 | 86.7 ± 21.2 |
| 300 | 97.6 ± 10.6 | 90.1 ± 18.2 | 97.5 ± 10.7 | 88.3 ± 18.4 |
| 325 | 98.5 ± 2.8 | 91.2 ± 17.2 | 96.8 ± 3.5 | 86.8 ± 16.6 |
| 350 | 98.8 ± 1.1 | 94.5 ± 10.6 | 96.1 ± 3.8 | 89.5 ± 10.6 |
| 375 | 98.4 ± 1.2 | 96.2 ± 2.5 | 95.6 ± 3.6 | 90.1 ± 6.7 |
| 400 | 97.9 ± 1.6 | 96.3 ± 2.5 | 93.1 ± 4.9 | 88.0 ± 7.6 |
| 450 | 97.3 ± 1.2 | 95.7 ± 1.7 | 91.8 ± 4.1 | 86.8 ± 6.2 |
| 500 | 99.9 ± 0.2 | 99.5 ± 0.6 | 99.4 ± 0.8 | 96.9 ± 3.4 |
| 550 | 100.0 ± 0.0 | 99.6 ± 0.4 | 99.5 ± 0.6 | 96.4 ± 3.3 |
| 600 | 99.9 ± 0.1 | 99.3 ± 0.8 | 99.4 ± 0.6 | 94.8 ± 5.9 |
| 650 | 99.8 ± 0.2 | 98.6 ± 1.1 | 97.5 ± 1.7 | 91.1 ± 6.2 |
| 700 | 99.8 ± 0.2 | 98.9 ± 1.0 | 97.9 ± 1.3 | 91.5 ± 7.9 |
| 750 | 99.8 ± 0.2 | 98.7 ± 1.2 | 97.5 ± 1.8 | 91.4 ± 6.3 |
| 800 | 100.0 ± 0.1 | 99.3 ± 0.9 | 99.2 ± 0.6 | 93.5 ± 6.4 |
| 850 | 100.0 ± 0.0 | 99.0 ± 1.3 | 99.4 ± 0.5 | 92.8 ± 8.6 |
| 900 | 100.0 ± 0.1 | 99.1 ± 1.1 | 99.1 ± 0.8 | 92.1 ± 7.0 |