Can I read Understanding the Effect of Sparsity on Neural Networks Robustness on EtoBox?

Understanding the Effect of Sparsity on Neural Networks Robustness by Timpl, Lukas; Entezari, Rahim; Sedghi, Hanie; Neyshabur, Behnam; Saukh, Olga is a scholarly article available to read on EtoBox.

What is Understanding the Effect of Sparsity on Neural Networks Robustness about?

This paper examines the impact of static sparsity on the robustness of a trained network to weight perturbations, data corruption, and adversarial examples. We show that, up to a certain sparsity achieved by increasing network width and depth while keeping the network capacity fixed, sparsified networks consistently match and often outperform their initially dense versions. Robustness and accuracy decline simultaneously for very high sparsity due to loose connectivity between network layers. Our findings show that a rapid robustness drop caused by network compression observed in the literature is due to a reduced network capacity rather than sparsity.

Author

Timpl, Lukas; Entezari, Rahim; Sedghi, Hanie; Neyshabur, Behnam; Saukh, Olga

Published

2022

Language

EN