Optimizing the Simplicial-Map Neural Network Architecture

Autor: Eduardo Paluzo-Hidalgo, Rocio Gonzalez-Diaz, Miguel A. Gutiérrez-Naranjo, Jónathan Heras
Jazyk: angličtina
Rok vydání: 2021
Předmět:
Zdroj: Journal of Imaging, Vol 7, Iss 9, p 173 (2021)
Druh dokumentu: article
ISSN: 2313-433X
DOI: 10.3390/jimaging7090173
Popis: Simplicial-map neural networks are a recent neural network architecture induced by simplicial maps defined between simplicial complexes. It has been proved that simplicial-map neural networks are universal approximators and that they can be refined to be robust to adversarial attacks. In this paper, the refinement toward robustness is optimized by reducing the number of simplices (i.e., nodes) needed. We have shown experimentally that such a refined neural network is equivalent to the original network as a classification tool but requires much less storage.
Databáze: Directory of Open Access Journals