Szczegóły publikacji
Opis bibliograficzny
Choice of the p-norm for high level classification features pruning in modern convolutional neural networks with local sensitivity analysis / Ernest JĘCZMIONEK, Piotr A. KOWALSKI // International Journal of Applied Mathematics and Computer Science ; ISSN 1641-876X. — 2023 — vol. 33 no. 4, s. 663–672. — Bibliogr. s. 670–672, Abstr. — Publikacja dostępna online od: 2023-12-21. — P. Kowalski - dod. afiliacja: Systems Research Institute Polish Academy of Sciences, Warsaw, Poland
Autorzy (2)
Słowa kluczowe
Dane bibliometryczne
| ID BaDAP | 151870 |
|---|---|
| Data dodania do BaDAP | 2024-02-09 |
| Tekst źródłowy | URL |
| DOI | 10.34768/amcs-2023-0047 |
| Rok publikacji | 2023 |
| Typ publikacji | artykuł w czasopiśmie |
| Otwarty dostęp |  |
| Creative Commons | |
| Czasopismo/seria | International Journal of Applied Mathematics and Computer Science |
Abstract
Transfer learning has surfaced as a compelling technique in machine learning, enabling the transfer of knowledge across networks. This study evaluates the efficacy of ImageNet pretrained state-of-the-art networks, including DenseNet, ResNet, and VGG, in implementing transfer learning for prepruned models on compact datasets, such as Fashion MNIST, CIFAR10, and CIFAR100. The primary objective is to reduce the number of neurons while preserving high-level features. To this end, local sensitivity analysis is employed alongside p-norms and various reduction levels. This investigation discovers that VGG16, a network rich in parameters, displays resilience to high-level feature pruning. Conversely, the ResNet architectures reveal an interesting pattern of increased volatility. These observations assist in identifying an optimal combination of the norm and the reduction level for each network architecture, thus offering valuable directions for model-specific optimization. This study marks a significant advance in understanding and implementing effective pruning strategies across diverse network architectures, paving the way for future research and applications. © 2023 E. Jeczmionek and P.A. Kowalski.
