s5_t2
done · peak unseen_mismatch 97.7
Referenced by
α schedule
α 5.0 ×19 α 5.0 ×19 α 5.0 ×19 α 5.0 ×19 α 5.0 ×19 α 5.0 ×19 α 5.0 ×19 α 5.0 ×19 α 5.0 ×19 α 5.0 ×19 α 5.0 ×19 α 5.0 ×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 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 ×19 α 1.2857 ×19 α 1.2857 ×19 α 1.2857 ×19 α 1.2857 ×19 α 1.2857 ×19 α 1.2857 ×19 α 1.2857 ×19 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 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 ± 1.5 | 49.9 ± 1.3 | 49.8 ± 1.9 | 49.6 ± 2.2 |
| 50 | 50.9 ± 1.8 | 50.0 ± 1.2 | 50.2 ± 1.6 | 50.4 ± 1.5 |
| 75 | 54.0 ± 1.9 | 51.6 ± 1.9 | 51.1 ± 1.7 | 50.1 ± 2.2 |
| 100 | 62.4 ± 6.2 | 58.5 ± 4.6 | 54.2 ± 4.3 | 51.8 ± 2.1 |
| 125 | 74.4 ± 8.7 | 70.9 ± 8.4 | 62.6 ± 8.9 | 59.0 ± 7.4 |
| 150 | 78.4 ± 4.5 | 76.4 ± 4.0 | 64.7 ± 6.3 | 61.9 ± 5.5 |
| 175 | 79.9 ± 3.9 | 78.6 ± 3.1 | 64.7 ± 5.8 | 62.6 ± 5.3 |
| 200 | 80.1 ± 3.3 | 78.6 ± 2.8 | 63.1 ± 4.8 | 61.8 ± 3.8 |
| 225 | 80.3 ± 3.3 | 78.9 ± 2.4 | 63.8 ± 4.3 | 62.7 ± 4.1 |
| 250 | 93.6 ± 4.5 | 92.3 ± 4.7 | 85.0 ± 10.0 | 83.4 ± 9.4 |
| 275 | 99.1 ± 0.7 | 98.6 ± 0.8 | 96.7 ± 2.0 | 95.3 ± 2.7 |
| 300 | 99.7 ± 0.2 | 99.5 ± 0.4 | 98.9 ± 0.8 | 97.7 ± 1.3 |
| 325 | 99.8 ± 0.2 | 99.5 ± 0.5 | 98.7 ± 0.8 | 97.0 ± 1.4 |
| 350 | 99.9 ± 0.1 | 99.5 ± 0.6 | 98.9 ± 1.1 | 96.7 ± 2.4 |
| 375 | 99.9 ± 0.2 | 99.4 ± 0.6 | 98.7 ± 0.9 | 96.2 ± 3.1 |
| 400 | 100.0 ± 0.0 | 99.6 ± 0.3 | 99.2 ± 0.7 | 97.1 ± 1.8 |
| 450 | 99.9 ± 0.1 | 99.7 ± 0.3 | 99.2 ± 0.9 | 96.9 ± 2.4 |
| 500 | 99.8 ± 0.2 | 98.9 ± 0.6 | 97.9 ± 1.5 | 93.1 ± 3.6 |
| 550 | 99.8 ± 0.1 | 99.2 ± 0.6 | 97.2 ± 1.3 | 93.2 ± 3.3 |
| 600 | 99.8 ± 0.2 | 98.9 ± 1.1 | 97.5 ± 1.4 | 91.8 ± 6.6 |
| 650 | 99.8 ± 0.2 | 99.1 ± 0.7 | 97.6 ± 1.6 | 91.6 ± 4.5 |
| 700 | 100.0 ± 0.0 | 98.5 ± 3.9 | 99.5 ± 0.5 | 93.9 ± 6.9 |
| 750 | 100.0 ± 0.0 | 99.0 ± 0.9 | 99.3 ± 0.7 | 92.2 ± 5.9 |
| 800 | 100.0 ± 0.0 | 98.9 ± 1.5 | 99.2 ± 0.7 | 92.0 ± 7.6 |
| 850 | 100.0 ± 0.0 | 99.1 ± 1.3 | 99.3 ± 0.6 | 92.2 ± 7.4 |
| 900 | 100.0 ± 0.0 | 99.1 ± 0.9 | 98.2 ± 1.3 | 90.9 ± 5.7 |