Skip to main navigation Skip to search Skip to main content

Purifying Adversarial Examples Using an Autoencoder

  • Thijs van Weezel (Corresponding author)
  • , Famke van Ree
  • , Tychon Bos
  • , Patrick Bastiaanssen
  • , Sibylle Hess

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

Abstract

One of the most prominent security challenges to neural networks are adversarial examples - inputs with often barely perceptible perturbations causing misclassification. In this study, we propose a defense mechanism that uses an autoencoder to restore adversarial examples before classification. That is, the autoencoder purifies input data points from potential adversarial perturbations. The method is titled Autoencoder-based Adversarial Purification (AAP). We demonstrate the effectiveness of AAP on multiple datasets, attack methods, and perturbation levels. While certain limitations exist, this research offers valuable insights and a promising direction for robust defense mechanisms in adversarial deep learning.

Original languageEnglish
Title of host publicationDiscovery Science - 27th International Conference, DS 2024, Proceedings
EditorsDino Pedreschi, Anna Monreale, Riccardo Guidotti, Francesca Naretto, Roberto Pellungrini
PublisherSpringer
Pages134-148
Number of pages15
ISBN (Electronic)9783031789809
ISBN (Print)9783031789793
DOIs
Publication statusPublished - 25 Jan 2025
Event27th International Conference on Discovery Science, DS 2024 - Pisa, Italy
Duration: 14 Oct 202416 Oct 2024

Publication series

NameLecture Notes in Computer Science
Volume15244
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349
NameLecture Notes in Artificial Intelligence
Volume15244

Conference

Conference27th International Conference on Discovery Science, DS 2024
Country/TerritoryItaly
CityPisa
Period14/10/2416/10/24

Bibliographical note

Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

Keywords

  • adversarial attack
  • autoencoder
  • purification

Fingerprint

Dive into the research topics of 'Purifying Adversarial Examples Using an Autoencoder'. Together they form a unique fingerprint.

Cite this