f14_d25_p600
done · peak unseen_mismatch 87.3
Referenced by
α schedule
α 1.4 ×450 α 1.0 ×150 α 1.4 ×450 α 1.0 ×150
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 |
|---|---|---|---|---|
| 150 | 52.8 ± 3.6 | 49.1 ± 1.9 | 52.6 ± 3.6 | 47.6 ± 3.0 |
| 300 | 84.8 ± 22.6 | 45.2 ± 19.5 | 84.4 ± 22.9 | 45.3 ± 19.0 |
| 450 | 100.0 ± 0.0 | 36.1 ± 30.8 | 100.0 ± 0.0 | 36.9 ± 30.6 |
| 600 | 100.0 ± 0.0 | 57.3 ± 31.9 | 100.0 ± 0.0 | 56.8 ± 31.4 |
| 750 | 100.0 ± 0.0 | 67.5 ± 28.8 | 100.0 ± 0.0 | 67.0 ± 29.1 |
| 900 | 99.9 ± 0.3 | 80.3 ± 31.2 | 100.0 ± 0.0 | 80.4 ± 31.1 |
| 1050 | 99.7 ± 0.6 | 86.7 ± 19.8 | 99.7 ± 0.6 | 85.9 ± 19.2 |
| 1200 | 100.0 ± 0.0 | 88.2 ± 14.7 | 100.0 ± 0.0 | 87.3 ± 15.2 |