Sparsity in neural networks can improve their privacy

Antoine Gonon, Léon Zheng, Clément Lalanne, Quoc-Tung Le, Guillaume Lauga +1 more
4/20/2023
cs.LGcs.AIcs.CR

No confident code implementation found for this paper. 1 possible match can be shown below.

Abstract

This article measures how sparsity can make neural networks more robust to membership inference attacks. The obtained empirical results show that sparsity improves the privacy of the network, while preserving comparable performances on the task at hand. This empirical study completes and extends existing literature.

View on arXivView PDF

Code Implementations

No confident code match yet

We couldn't find an author-owned or strongly-evidenced community implementation for this paper. 1 weaker match is hidden by default — verify before relying on them.

No code implementations found yet.

Know of an implementation? Let us know in the comments below!

Similar Papers

Can sparsity improve the privacy of neural networks?

92% match2023

Spurious Privacy Leakage in Neural Networks

88% match2025

On the Privacy Risks of Spiking Neural Networks: A Membership Inference Analysis

86% match1 repo42025

Discriminative Adversarial Privacy: Balancing Accuracy and Membership Privacy in Neural Networks

85% match2023

Learnability and Privacy Vulnerability are Entangled in a Few Critical Weights

85% match2026

Impact of Dataset Properties on Membership Inference Vulnerability of Deep Transfer Learning

85% match1 repo12024

Cite this paper

@article{gonon2023sparsity,
  title  = {Sparsity in neural networks can improve their privacy},
  author = {Antoine Gonon and Léon Zheng and Clément Lalanne and Quoc-Tung Le and Guillaume Lauga and Can Pouliquen},
  year   = {2023},
  eprint = {2304.10553},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2304.10553v2}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2304.10553.svg)](https://codeofpaper.com/paper/2304.10553)

Discussion