smooth_K48_cont
done · peak unseen_mismatch 91.3
Referenced by
α schedule
α 2.9821 ×19 α 2.9464 ×19 α 2.9107 ×19 α 2.875 ×19 α 2.8393 ×19 α 2.8036 ×19 α 2.7679 ×19 α 2.7321 ×19 α 2.6964 ×19 α 2.6607 ×19 α 2.625 ×19 α 2.5893 ×19 α 2.5536 ×19 α 2.5179 ×19 α 2.4821 ×19 α 2.4464 ×19 α 2.4107 ×19 α 2.375 ×19 α 2.3393 ×19 α 2.3036 ×19 α 2.2679 ×19 α 2.2321 ×19 α 2.1964 ×19 α 2.1607 ×19 α 2.125 ×19 α 2.0893 ×19 α 2.0536 ×19 α 2.0178 ×19 α 1.9821 ×19 α 1.9464 ×19 α 1.9107 ×19 α 1.875 ×19 α 1.8393 ×19 α 1.8036 ×19 α 1.7678 ×19 α 1.7321 ×19 α 1.6964 ×18 α 1.6607 ×18 α 1.625 ×18 α 1.5893 ×18 α 1.5536 ×18 α 1.5178 ×18 α 1.4821 ×18 α 1.4464 ×18 α 1.4107 ×18 α 1.375 ×18 α 1.3393 ×18 α 1.3036 ×18
Figures
Results
accuracy by training iteration, mean ± sd over seeds · schedule ends at α = 1.3036
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 | 51.4 ± 1.5 | 49.5 ± 1.6 | 49.9 ± 1.8 | 49.7 ± 2.0 |
| 50 | 50.9 ± 1.7 | 49.5 ± 1.4 | 50.0 ± 1.5 | 50.8 ± 1.3 |
| 75 | 52.0 ± 1.9 | 50.0 ± 1.8 | 50.5 ± 1.7 | 50.1 ± 1.9 |
| 100 | 52.9 ± 3.0 | 51.7 ± 2.6 | 51.2 ± 1.6 | 49.8 ± 1.2 |
| 125 | 64.2 ± 11.1 | 58.5 ± 7.7 | 58.6 ± 10.0 | 54.2 ± 4.9 |
| 150 | 79.0 ± 12.8 | 71.5 ± 13.0 | 72.0 ± 14.4 | 64.8 ± 12.5 |
| 175 | 87.7 ± 10.6 | 81.9 ± 11.0 | 80.5 ± 12.5 | 73.8 ± 11.5 |
| 200 | 90.5 ± 9.7 | 86.3 ± 9.0 | 82.8 ± 10.6 | 77.1 ± 9.6 |
| 225 | 91.5 ± 8.6 | 89.8 ± 8.1 | 83.2 ± 9.9 | 79.6 ± 9.9 |
| 250 | 92.5 ± 4.4 | 90.4 ± 5.2 | 82.8 ± 6.4 | 78.9 ± 6.8 |
| 275 | 93.9 ± 2.4 | 92.0 ± 2.6 | 83.5 ± 6.3 | 79.6 ± 6.3 |
| 300 | 94.2 ± 1.5 | 92.3 ± 1.8 | 84.0 ± 5.2 | 79.8 ± 5.4 |
| 325 | 94.8 ± 1.6 | 92.7 ± 2.0 | 84.0 ± 4.7 | 78.9 ± 5.4 |
| 350 | 94.7 ± 1.2 | 92.8 ± 1.5 | 83.1 ± 3.6 | 78.0 ± 4.2 |
| 375 | 94.9 ± 1.2 | 93.1 ± 1.8 | 84.4 ± 3.4 | 79.8 ± 4.7 |
| 400 | 95.8 ± 1.3 | 93.8 ± 1.5 | 84.8 ± 2.6 | 79.8 ± 3.9 |
| 450 | 96.0 ± 1.2 | 94.3 ± 1.4 | 86.0 ± 3.9 | 80.5 ± 4.9 |
| 500 | 96.8 ± 0.7 | 94.8 ± 1.2 | 88.2 ± 2.5 | 81.0 ± 4.6 |
| 550 | 97.8 ± 0.7 | 95.8 ± 0.9 | 89.9 ± 2.3 | 84.1 ± 3.3 |
| 600 | 98.1 ± 0.6 | 96.6 ± 1.0 | 92.0 ± 2.1 | 85.2 ± 3.7 |
| 650 | 98.5 ± 0.4 | 97.1 ± 0.8 | 92.3 ± 2.7 | 84.9 ± 5.1 |
| 700 | 99.1 ± 0.6 | 97.7 ± 1.1 | 94.0 ± 2.9 | 87.0 ± 5.0 |
| 750 | 99.4 ± 0.3 | 97.8 ± 1.4 | 95.2 ± 2.0 | 87.4 ± 5.7 |
| 800 | 99.6 ± 0.3 | 98.4 ± 1.2 | 96.7 ± 1.5 | 89.0 ± 5.3 |
| 850 | 99.7 ± 0.2 | 98.8 ± 0.9 | 97.5 ± 1.6 | 90.3 ± 5.3 |
| 900 | 99.8 ± 0.1 | 99.1 ± 0.5 | 97.8 ± 1.4 | 91.3 ± 4.4 |