Pré-Publication, Document De Travail Année : 2025

On the impact of the parametrization of deep convolutional neural networks on post-training quantization

Résumé

This paper introduces novel theoretical approximation bounds for the output of quantized neural networks, with a focus on convolutional neural networks (CNN). By considering layerwise parametrization and focusing on the quantization of weights, we provide bounds that gain several orders of magnitude compared to state-of-the art results on classical deep convolutional neural netorks such as MobileNetV2 or ResNets. These gains are achieved by improving the behaviour of the approximation bounds with respect to the depth parameter, which has the most impact on the approximation error induced by quantization. To complement our theoretical result, we provide a numerical exploration of our bounds on Mo-bileNetV2 and ResNets.

Fichier principal
Vignette du fichier
main_hal.pdf (947.27 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Licence

Dates et versions

hal-04922698 , version 1 (31-01-2025)
hal-04922698 , version 2 (09-10-2025)

Licence

Identifiants

  • HAL Id : hal-04922698 , version 1

Citer

Samy Houache, Jean François Aujol, Yann Traonmilin. On the impact of the parametrization of deep convolutional neural networks on post-training quantization. 2025. ⟨hal-04922698v1⟩
146 Consultations
425 Téléchargements

Partager

  • More