s4_t4
done · peak unseen_mismatch 98.6
Referenced by
α schedule
α 4.0 ×19 α 4.0 ×19 α 4.0 ×19 α 4.0 ×19 α 4.0 ×19 α 4.0 ×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 3.0952 ×19 α 3.0952 ×19 α 3.0952 ×19 α 3.0952 ×19 α 3.0952 ×19 α 3.0952 ×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 2.1905 ×19 α 2.1905 ×19 α 2.1905 ×19 α 2.1905 ×19 α 2.1905 ×19 α 2.1905 ×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 ×18 α 1.2857 ×18 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
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 | 51.5 ± 1.6 | 49.7 ± 1.6 | 49.9 ± 2.0 | 49.7 ± 2.1 |
| 50 | 51.1 ± 1.7 | 49.9 ± 1.3 | 50.1 ± 1.5 | 50.4 ± 1.5 |
| 75 | 52.9 ± 2.1 | 50.9 ± 1.9 | 50.6 ± 1.7 | 50.0 ± 2.1 |
| 100 | 58.7 ± 5.5 | 55.5 ± 4.2 | 52.5 ± 4.0 | 51.1 ± 2.2 |
| 125 | 69.9 ± 12.1 | 63.5 ± 9.7 | 63.2 ± 12.4 | 57.4 ± 8.3 |
| 150 | 87.8 ± 12.6 | 75.4 ± 15.4 | 85.3 ± 14.2 | 72.1 ± 16.1 |
| 175 | 97.5 ± 9.2 | 75.3 ± 21.3 | 97.3 ± 9.6 | 73.3 ± 21.7 |
| 200 | 98.3 ± 7.3 | 74.5 ± 28.0 | 98.4 ± 7.1 | 73.4 ± 28.3 |
| 225 | 100.0 ± 0.0 | 78.9 ± 24.7 | 100.0 ± 0.0 | 77.9 ± 24.6 |
| 250 | 99.6 ± 0.8 | 89.3 ± 18.4 | 98.8 ± 2.0 | 86.5 ± 18.8 |
| 275 | 98.5 ± 1.2 | 92.9 ± 11.5 | 96.5 ± 3.1 | 88.6 ± 11.8 |
| 300 | 96.8 ± 2.3 | 92.0 ± 10.4 | 92.9 ± 5.6 | 86.0 ± 11.3 |
| 325 | 95.6 ± 2.8 | 93.3 ± 3.9 | 90.3 ± 6.3 | 85.2 ± 8.2 |
| 350 | 99.0 ± 1.0 | 97.8 ± 1.7 | 97.2 ± 2.7 | 94.6 ± 3.6 |
| 375 | 100.0 ± 0.1 | 99.3 ± 1.8 | 99.7 ± 0.5 | 98.3 ± 2.8 |
| 400 | 99.9 ± 0.2 | 99.7 ± 0.4 | 99.6 ± 0.6 | 98.2 ± 2.3 |
| 450 | 100.0 ± 0.1 | 99.6 ± 0.4 | 99.5 ± 0.5 | 97.5 ± 2.2 |
| 500 | 97.5 ± 1.1 | 95.7 ± 1.6 | 92.0 ± 3.6 | 86.7 ± 4.5 |
| 550 | 97.1 ± 0.7 | 95.2 ± 1.5 | 89.8 ± 2.6 | 84.9 ± 3.7 |
| 600 | 100.0 ± 0.1 | 99.8 ± 0.4 | 99.9 ± 0.3 | 98.6 ± 2.4 |
| 650 | 99.9 ± 0.1 | 99.5 ± 0.6 | 99.6 ± 0.5 | 96.4 ± 3.1 |
| 700 | 99.7 ± 0.2 | 98.3 ± 1.6 | 96.8 ± 1.6 | 89.2 ± 6.0 |
| 750 | 99.8 ± 0.2 | 98.6 ± 1.2 | 97.7 ± 1.7 | 91.2 ± 6.2 |
| 800 | 99.9 ± 0.1 | 99.1 ± 0.9 | 98.6 ± 1.1 | 92.5 ± 4.8 |
| 850 | 100.0 ± 0.1 | 99.3 ± 0.7 | 99.2 ± 0.7 | 92.8 ± 5.5 |
| 900 | 100.0 ± 0.1 | 98.8 ± 1.9 | 99.0 ± 1.3 | 91.0 ± 8.0 |