Table 3 Accuracy, \(\kappa \), and % of cases where each algorithm outperformed all other algorithms for bespoke CNN
Algorithm | Average ± standard deviation % accuracy | Average ± standard deviation Cohen’s \(\kappa \) | % of cases where algorithm was best |
|---|---|---|---|
Head Re-train | 76.4 ± 7.1 | 0.666 ± 0.095 | 22.2 |
Subspace alignment | 45.4 ± 13.5 | 0.220 ± 0.163 | 0.7 |
Per-Class CORAL | 76.5 ± 6.2 | 0.670 ± 0.083 | 9.0 |
CORAL | 76.7 ± 6.2 | 0.671 ± 0.084 | 14.6 |
DDC | 77.4 ± 7.9 | 0.683 ± 0.104 | 52.1 |