Delving Deep into Interpreting Neural Nets with Piece-Wise Affine Representation - Sorbonne Université
Communication Dans Un Congrès Année : 2019

Delving Deep into Interpreting Neural Nets with Piece-Wise Affine Representation

Résumé

Deep convolutional neural networks (CNNs) are now ubiquitous in computer vision problems. However, these models usually describe very complicated functions of the input images. For a number of application, it is of utmost importance to be able to explain the decisions of a network, e.g. by highlighting the most relevant pixels in an image or a feature map w.r.t. a particular class. In this paper, we show that CNNs locally describe piece-wise affine functions of each pixel, whose coefficient and bias can be retrieved analytically. We apply our methodology on several popular CNNs and draw interesting conclusions on the relative contributions of pixels and biases for these networks.
Fichier principal
Vignette du fichier
ICIP2019.pdf (597.8 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-02507512 , version 1 (13-03-2020)

Identifiants

Citer

Yifu Chen, Antoine Saporta, Arnaud Dapogny, Matthieu Cord. Delving Deep into Interpreting Neural Nets with Piece-Wise Affine Representation. 2019 IEEE International Conference on Image Processing (ICIP), Sep 2019, Taipei, Taiwan. pp.609-613, ⟨10.1109/ICIP.2019.8803817⟩. ⟨hal-02507512⟩
109 Consultations
178 Téléchargements

Altmetric

Partager

More