Characterizing and Measuring the Similarity of Neural Networks with Persistent Homology

Asier Gutiérrez-Fandiño, David Pérez-Fernández, Jordi Armengol-Estapé, Marta Villegas

Publication date: 19 January 2021

Abstract: Characterizing the structural properties of neural networks is crucial yet poorly understood, and there are no well-established similarity measures between networks. In this work, we observe that neural networks can be represented as abstract simplicial complex and analyzed using their topological 'fingerprints' via Persistent Homology (PH). We then describe a PH-based representation proposed for characterizing and measuring similarity of neural networks. We empirically show the effectiveness of this representation as a descriptor of different architectures in several datasets. This approach based on Topological Data Analysis is a step towards better understanding neural networks and serves as a useful similarity measure.

Has companion code repository: https://github.com/asier-gutierrez/nn-similarity

This page was built for publication: Characterizing and Measuring the Similarity of Neural Networks with Persistent Homology

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6358605)