A method for mapping Keras convolutional neural network models for classification to a neuron-level weighted graph
MethodsX · 2026-12 · 17 · 104243
DOI: 10.1016/j.mex.2026.104243 ↗Kurz gefasst
Fachartikel in der Zeitschrift MethodsX, 2026-12. Verfasst von Markus Schatten und 1 weiteren Autor:innen. Beteiligte Einrichtung: Faculty of Organization and Informatics, University of Zagreb, Artificial Intelligence Laboratory, Pavlinska 2, Varaždin, 42000, Croatia. Die Arbeit nennt 23 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: European Commission.
Abstract (Originalsprache)
This study proposes and implements a method for mapping supported Keras convolutional neural network models for classification to a weighted graph at the level of individual neurons. The reviewed literature lacks a systematic procedure covering the diverse connectivity patterns found in convolutional architectures. Mapping rules are defined for standard, grouped, dilated, depthwise, and separable convolutional layers, as well as pooling layers and layers of the classification head. The implementation is verified using structural invariants and validated for topology and weights using small reference models. Practical applicability and scalability are evaluated on models based on AlexNet, VGG11, and MobileNetV1 at two input sizes, and a selected set of complex-network metrics is computed for the resulting graphs. The results show that the method consistently maps supported Keras convolutional neural network models to large weighted graphs suitable for further topological analysis.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Faculty of Organization and Informatics, University of Zagreb, Artificial Intelligence Laboratory, Pavlinska 2, Varaždin, 42000, Croatia.
Förderung
- European Commission
Zitieren
APA
Schatten, M., & Kurtanović, A. (2026). A method for mapping Keras convolutional neural network models for classification to a neuron-level weighted graph. MethodsX, 17, 104243. https://doi.org/10.1016/j.mex.2026.104243
Vancouver
Schatten M, Kurtanović A. A method for mapping Keras convolutional neural network models for classification to a neuron-level weighted graph. MethodsX. 2026;17:104243. doi:10.1016/j.mex.2026.104243
BibTeX
@article{Schatten2026,
author = {Schatten, Markus and Kurtanović, Admir},
title = {A method for mapping Keras convolutional neural network models for classification to a neuron-level weighted graph},
journal = {MethodsX},
year = {2026},
volume = {17},
pages = {104243},
doi = {10.1016/j.mex.2026.104243}
}
EUVIMED