Deep neural rejection against adversarial examples

Angelo Sotgiu
First
;
Ambra Demontis
;
Marco Melis;Battista Biggio;Giorgio Fumera;Fabio Roli
Last
2020

Abstract

Despite the impressive performances reported by deep neural networks in different application domains, they remain largely vulnerable to adversarial examples, i.e., input samples that are carefully perturbed to cause misclassification at test time. In this work, we propose a deep neural rejection mechanism to detect adversarial examples, based on the idea of rejecting samples that exhibit anomalous feature representations at different network layers. With respect to competing approaches, our method does not require generating adversarial examples at training time, and it is less computationally demanding. To properly evaluate our method, we define an adaptive white-box attack that is aware of the defense mechanism and aims to bypass it. Under this worst-case setting, we empirically show that our approach outperforms previously proposed methods that detect adversarial examples by only analyzing the feature representation provided by the output network layer.
Inglese
2020
1
10
10
https://link.springer.com/article/10.1186/s13635-020-00105-y?wt_mc=Internal.Event.1.SEM.ArticleAuthorIncrementalIssue
Esperti anonimi
internazionale
scientifica
Adversarial machine Learning; Deep neural network;, Adversarial examples; Computer Science - Computer Vision and Pattern Recognition; Computer Science - Computer Vision and Pattern Recognition; Computer Science - Learning; 68T45
Article number: 5 (2020)
sì
Sotgiu, Angelo; Demontis, Ambra; Melis, Marco; Biggio, Battista; Fumera, Giorgio; Feng, Xiaoyi; Roli, Fabio
1.1 Articolo in rivista
info:eu-repo/semantics/article
1 Contributo su Rivista::1.1 Articolo in rivista
262
7
open
Files in This Item:
File Size Format  
Sotgiu20.pdf

open access

Description: articolo
Type: versione editoriale
Size 3.6 MB
Format Adobe PDF
3.6 MB Adobe PDF View/Open

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

Questionnaire and social

Share on:
Impostazioni cookie