Enhanced network compression through tensor decompositions and pruning - Archive ouverte HAL Accéder directement au contenu
Article Dans Une Revue IEEE Transactions on Neural Networks and Learning Systems Année : 2024

Enhanced network compression through tensor decompositions and pruning

Résumé

Network compression techniques that combine tensor decompositions and pruning have shown promise in leveraging the advantages of both strategies. In this work, we propose NORTON (enhanced Network cOmpRession through TensOr decompositions and pruNing), a novel method for network compression. NORTON introduces the concept of filter decomposition, enabling a more detailed decomposition of the network while preserving the weight's multidimensional properties. Our method incorporates a novel structured pruning approach, effectively integrating the decomposed model. Through extensive experiments on various architectures, benchmark datasets, and representative vision tasks, we demonstrate the usefulness of our method. NORTON achieves superior results compared to stateof-the-art techniques in terms of complexity and accuracy. Our code is also available for research purposes.
Fichier principal
Vignette du fichier
TNNLS-2023-P-29253-1.pdf (3.05 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04475167 , version 1 (23-02-2024)

Identifiants

Citer

Van Tien Pham, Yassine Zniyed, Thanh Phuong Nguyen. Enhanced network compression through tensor decompositions and pruning. IEEE Transactions on Neural Networks and Learning Systems, In press, ⟨10.1109/TNNLS.2024.3370294⟩. ⟨hal-04475167⟩
144 Consultations
77 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More