arXiv Open Access 2024

Exploring connections of spectral analysis and transfer learning in medical imaging

Yucheng Lu Dovile Juodelyte Jonathan D. Victor Veronika Cheplygina
Lihat Sumber

Abstrak

In this paper, we use spectral analysis to investigate transfer learning and study model sensitivity to frequency shortcuts in medical imaging. By analyzing the power spectrum density of both pre-trained and fine-tuned model gradients, as well as artificially generated frequency shortcuts, we observe notable differences in learning priorities between models pre-trained on natural vs medical images, which generally persist during fine-tuning. We find that when a model's learning priority aligns with the power spectrum density of an artifact, it results in overfitting to that artifact. Based on these observations, we show that source data editing can alter the model's resistance to shortcut learning.

Topik & Kata Kunci

Penulis (4)

Y

Yucheng Lu

D

Dovile Juodelyte

J

Jonathan D. Victor

V

Veronika Cheplygina

Format Sitasi

Lu, Y., Juodelyte, D., Victor, J.D., Cheplygina, V. (2024). Exploring connections of spectral analysis and transfer learning in medical imaging. https://arxiv.org/abs/2407.11379

Akses Cepat

Lihat di Sumber
Informasi Jurnal
Tahun Terbit
2024
Bahasa
en
Sumber Database
arXiv
Akses
Open Access ✓