free-00c20121
done · peak unseen_mismatch 90.8
Referenced by
α schedule
α 4.4 ×10 α 4.5 ×10 α 4.7 ×10 α 4.9 ×10 α 4.7 ×10 α 5.0 ×10 α 4.8 ×10 α 4.9 ×10 α 4.8 ×10 α 4.7 ×20 α 4.9 ×10 α 4.7 ×10 α 4.9 ×10 α 4.7 ×10 α 4.9 ×10 α 0.8 ×10 α 0.9 ×10 α 0.8 ×10 α 1.0 ×10 α 0.5 ×10 α 1.0 ×10 α 0.7 ×30 α 0.9 ×20 α 0.8 ×10 α 0.9 ×120
Figures
Results
accuracy by training iteration, mean ± sd over seeds · schedule ends at α = 0.9
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 |
|---|---|---|---|---|
| 100 | 60.8 ± 7.7 | 57.1 ± 6.2 | 54.4 ± 5.8 | 51.7 ± 3.5 |
| 110 | 65.7 ± 9.1 | 62.0 ± 7.8 | 56.7 ± 7.5 | 53.6 ± 5.3 |
| 120 | 70.3 ± 9.8 | 67.0 ± 8.7 | 59.6 ± 8.7 | 56.1 ± 6.6 |
| 130 | 74.4 ± 8.4 | 71.1 ± 7.5 | 61.7 ± 8.6 | 58.1 ± 6.7 |
| 140 | 77.2 ± 7.3 | 74.3 ± 6.7 | 63.1 ± 8.4 | 59.9 ± 6.6 |
| 150 | 78.7 ± 5.6 | 76.1 ± 5.0 | 63.5 ± 7.5 | 60.7 ± 6.1 |
| 160 | 79.4 ± 5.0 | 77.2 ± 4.3 | 63.8 ± 7.0 | 61.0 ± 5.8 |
| 170 | 83.0 ± 6.4 | 80.5 ± 5.9 | 69.3 ± 9.2 | 66.3 ± 8.0 |
| 180 | 90.2 ± 7.8 | 87.7 ± 7.7 | 82.5 ± 11.1 | 78.8 ± 10.9 |
| 190 | 95.1 ± 7.2 | 92.7 ± 7.6 | 91.3 ± 9.7 | 88.0 ± 10.1 |
| 200 | 96.8 ± 6.3 | 94.4 ± 7.3 | 94.3 ± 8.5 | 90.5 ± 9.2 |
| 210 | 97.4 ± 5.7 | 93.8 ± 8.2 | 95.8 ± 7.5 | 90.8 ± 9.3 |
| 220 | 97.4 ± 3.3 | 83.0 ± 12.2 | 96.6 ± 4.6 | 80.5 ± 12.1 |
| 230 | 98.7 ± 2.5 | 75.6 ± 19.1 | 98.2 ± 3.8 | 73.0 ± 19.3 |
| 240 | 99.7 ± 0.5 | 64.4 ± 20.1 | 99.8 ± 0.4 | 62.2 ± 19.0 |
| 250 | 100.0 ± 0.0 | 66.5 ± 21.1 | 100.0 ± 0.1 | 63.5 ± 21.4 |
| 260 | 100.0 ± 0.0 | 72.5 ± 13.8 | 100.0 ± 0.0 | 70.6 ± 13.4 |
| 270 | 100.0 ± 0.0 | 85.4 ± 6.8 | 100.0 ± 0.0 | 83.0 ± 6.2 |
| 280 | 100.0 ± 0.0 | 87.5 ± 5.9 | 100.0 ± 0.0 | 85.1 ± 4.9 |
| 290 | 100.0 ± 0.0 | 88.4 ± 8.2 | 100.0 ± 0.0 | 86.3 ± 6.6 |
| 300 | 100.0 ± 0.0 | 68.2 ± 9.9 | 99.9 ± 0.2 | 66.8 ± 7.4 |
| 310 | 100.0 ± 0.0 | 76.8 ± 15.4 | 100.0 ± 0.0 | 75.5 ± 15.4 |
| 320 | 100.0 ± 0.0 | 85.1 ± 18.2 | 100.0 ± 0.0 | 84.2 ± 19.3 |
| 330 | 100.0 ± 0.0 | 84.9 ± 0.0 | 100.0 ± 0.0 | 85.1 ± 0.0 |
| 340 | 100.0 ± 0.0 | 74.9 ± 0.0 | 100.0 ± 0.0 | 73.4 ± 0.0 |
| 350 | 100.0 ± 0.0 | 60.8 ± 0.0 | 100.0 ± 0.0 | 59.1 ± 0.0 |
| 360 | 100.0 ± 0.0 | 59.4 ± 0.0 | 100.0 ± 0.0 | 55.9 ± 0.0 |
| 370 | 100.0 ± 0.0 | 76.1 ± 0.0 | 100.0 ± 0.0 | 76.4 ± 0.0 |
| 380 | 100.0 ± 0.0 | 77.1 ± 0.0 | 100.0 ± 0.0 | 75.9 ± 0.0 |
| 390 | 100.0 ± 0.0 | 55.2 ± 0.0 | 100.0 ± 0.0 | 50.8 ± 0.0 |
| 400 | 100.0 ± 0.0 | 87.7 ± 0.0 | 100.0 ± 0.0 | 88.0 ± 0.0 |