Can Generalised Divergences Help for Invariant Neural Networks? - Morphologie mathématique (CMM) Access content directly
Book Sections Year : 2023

Can Generalised Divergences Help for Invariant Neural Networks?

Abstract

We consider a framework including multiple augmentation regularisation by generalised divergences to induce invariance for nongroup transformations during training of convolutional neural networks. Experiments on supervised classification of images at different scales not considered during training illustrate that our proposed method performs better than classical data augmentation.
Fichier principal
Vignette du fichier
OrbitRegularization.pdf (202.99 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-04303522 , version 1 (23-11-2023)

Identifiers

Cite

Santiago Velasco-Forero. Can Generalised Divergences Help for Invariant Neural Networks?. Geometric Science of Information, 14071, Springer Nature Switzerland, pp.82-90, 2023, Lecture Notes in Computer Science, ⟨10.1007/978-3-031-38271-0_9⟩. ⟨hal-04303522⟩
7 View
14 Download

Altmetric

Share

Gmail Facebook X LinkedIn More