Table 2 Performance comparison among different loss functions.
From: Two stage multiobjective topology optimization method via SwinUnet with enhanced generalization
Loss function | \(E_c\) | \(E_{\sigma _{PN}}\) | \(E_v\) | BSR | SSR | OSR |
|---|---|---|---|---|---|---|
\(\mathscr {L}_{Dice}\) | \(-\)6.15% | \(-\)1.86% | 0.01% | 4.85% | 8.18% | 58.79% |
\(\mathscr {L}_{BCE}\) | \(-\)2.52% | 0.63% | \(-\)0.55% | 7.27% | 16.06% | 53.33% |
\(\mathscr {L}_{B \& D}\) | \(-\)5.22% | \(-\)1.25% | \(-\)0.33% | 6.52% | 10.30% | 54.54% |
