Searching for N:M Fine-grained Sparsity of Weights and Activations in Neural Networks

Ruth Akiva-Hochman, Shahaf E. Finder, Javier S. Turek, Eran Treister

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Scopus citations

Abstract

Sparsity in deep neural networks has been extensively studied to compress and accelerate models for environments with limited resources. The general approach of pruning aims at enforcing sparsity on the obtained model, with minimal accuracy loss, but with a sparsity structure that enables acceleration on hardware. The sparsity can be enforced on either the weights or activations of the network, and existing works tend to focus on either one for the entire network. In this paper, we suggest a strategy based on Neural Architecture Search (NAS) to sparsify both activations and weights throughout the network, while utilizing the recent approach of N:M fine-grained structured sparsity that enables practical acceleration on dedicated GPUs. We show that a combination of weight and activation pruning is superior to each option separately. Furthermore, during the training, the choice between pruning the weights of activations can be motivated by practical inference costs (e.g., memory bandwidth). We demonstrate the efficiency of the approach on several image classification datasets.

Original languageEnglish
Title of host publicationComputer Vision – ECCV 2022 Workshops, Proceedings
EditorsLeonid Karlinsky, Tomer Michaeli, Ko Nishino
PublisherSpringer Science and Business Media Deutschland GmbH
Pages130-143
Number of pages14
ISBN (Print)9783031250811
DOIs
StatePublished - 1 Jan 2023
Event17th European Conference on Computer Vision, ECCV 2022 - Tel Aviv, Israel
Duration: 23 Oct 202227 Oct 2022

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume13807 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference17th European Conference on Computer Vision, ECCV 2022
Country/TerritoryIsrael
CityTel Aviv
Period23/10/2227/10/22

Keywords

  • Activation pruning
  • N:M fine-grained Sparsity
  • Neural architecture search
  • Weight pruning

ASJC Scopus subject areas

  • Theoretical Computer Science
  • General Computer Science

Fingerprint

Dive into the research topics of 'Searching for N:M Fine-grained Sparsity of Weights and Activations in Neural Networks'. Together they form a unique fingerprint.

Cite this