r8_c3
done · peak unseen_mismatch 99.8
Referenced by
α schedule
α 5.0 ×25 α 5.0 ×25 α 5.0 ×25 α 1.0634 ×25 α 1.0536 ×25 α 5.0 ×25 α 5.0 ×25 α 5.0 ×25 α 0.9043 ×25 α 0.8215 ×25 α 0.7122 ×25 α 3.084 ×25 α 2.1282 ×25 α 0.1755 ×25
Figures
Results
accuracy by training iteration, mean ± sd over seeds · schedule ends at α = 0.1755
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 | 50.9 ± 1.3 | 50.5 ± 1.6 | 49.9 ± 1.4 | 49.7 ± 2.0 |
| 50 | 51.5 ± 2.0 | 50.5 ± 1.5 | 50.0 ± 1.3 | 50.0 ± 1.6 |
| 75 | 53.4 ± 2.6 | 50.7 ± 1.7 | 50.4 ± 1.3 | 49.6 ± 1.5 |
| 100 | 56.0 ± 8.5 | 52.6 ± 3.2 | 52.8 ± 6.4 | 50.3 ± 1.8 |
| 125 | 64.7 ± 16.2 | 54.8 ± 8.7 | 61.5 ± 16.3 | 52.7 ± 7.4 |
| 150 | 78.7 ± 16.6 | 71.3 ± 14.2 | 74.3 ± 17.9 | 66.3 ± 13.9 |
| 175 | 87.4 ± 9.7 | 83.2 ± 9.7 | 81.1 ± 12.3 | 76.4 ± 11.8 |
| 200 | 89.4 ± 5.5 | 86.7 ± 5.6 | 80.9 ± 9.4 | 77.6 ± 8.9 |
| 225 | 99.1 ± 1.1 | 95.3 ± 11.2 | 98.2 ± 2.3 | 93.4 ± 12.3 |
| 250 | 100.0 ± 0.1 | 98.2 ± 3.8 | 99.9 ± 0.3 | 97.3 ± 4.8 |
| 275 | 100.0 ± 0.0 | 97.7 ± 7.3 | 100.0 ± 0.1 | 97.3 ± 7.2 |
| 300 | 97.0 ± 2.6 | 94.7 ± 3.7 | 91.3 ± 6.7 | 87.4 ± 8.6 |
| 325 | 98.1 ± 1.2 | 95.7 ± 2.3 | 94.1 ± 3.3 | 89.4 ± 5.7 |
| 350 | 100.0 ± 0.0 | 99.9 ± 0.2 | 100.0 ± 0.0 | 99.8 ± 0.3 |