Learning Sparse Neural Networks via Sensitivity-Driven Regularization - Equipe Multimédia Accéder directement au contenu
Communication Dans Un Congrès Année : 2018

Learning Sparse Neural Networks via Sensitivity-Driven Regularization

Résumé

The ever-increasing number of parameters in deep neural networks poses challenges for memory-limited applications. Regularize-and-prune methods aim at meeting these challenges by sparsifying the network weights. In this context we quantify the output sensitivity to the parameters (i.e. their relevance to the network output) and introduce a regularization term that gradually lowers the absolute value of parameters with low sensitivity. Thus, a very large fraction of the parameters approach zero and are eventually set to zero by simple thresholding. Our method surpasses most of the recent techniques both in terms of sparsity and error rates. In some cases, the method reaches twice the sparsity obtained by other techniques at equal error rates.
Fichier principal
Vignette du fichier
7644-learning-sparse-neural-networks-via-sensitivity-driven-regularization.pdf (367.63 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01995794 , version 1 (27-01-2019)

Identifiants

  • HAL Id : hal-01995794 , version 1

Citer

Enzo Tartaglione, Skjalg Lepsoy, Attilio Fiandrotti, Gianluca Francini. Learning Sparse Neural Networks via Sensitivity-Driven Regularization. The Thirty-second Conference on Neural Information Processing Systems (NIPS- NeurIPS 2018 ), Dec 2018, Montréal, Canada. ⟨hal-01995794⟩
109 Consultations
101 Téléchargements

Partager

Gmail Facebook X LinkedIn More