Deep generative models are rapidly becoming a common tool for researchers and developers. However, as exhaustively shown for the family of discriminative models, the test-time inference of deep neural networks cannot be fully controlled and erroneous behaviors can be induced by an attacker. In the present work, we show how a malicious user can force a pre-trained generator to reproduce arbitrary data instances by feeding it suitable adversarial inputs. Moreover, we show that these adversarial latent vectors can be shaped so as to be statistically indistinguishable from the set of genuine inputs. The proposed attack technique is evaluated with respect to various GAN images generators using different architectures, training processes and for both conditional and not-conditional setups.

Adversarial out-domain examples for generative models / Pasquini, Dario; Mingione, Marco; Bernaschi, Massimo. - (2019), pp. 272-280. (Intervento presentato al convegno MaL2CSec 2019 : Workshop on Machine Learning for Cyber-Crime Investigation and Cybersecurity tenutosi a Stockholm; Sweden) [10.1109/EuroSPW.2019.00037].

Adversarial out-domain examples for generative models

PASQUINI, DARIO
;
MINGIONE, MARCO;Massimo Bernaschi
2019

Abstract

Deep generative models are rapidly becoming a common tool for researchers and developers. However, as exhaustively shown for the family of discriminative models, the test-time inference of deep neural networks cannot be fully controlled and erroneous behaviors can be induced by an attacker. In the present work, we show how a malicious user can force a pre-trained generator to reproduce arbitrary data instances by feeding it suitable adversarial inputs. Moreover, we show that these adversarial latent vectors can be shaped so as to be statistically indistinguishable from the set of genuine inputs. The proposed attack technique is evaluated with respect to various GAN images generators using different architectures, training processes and for both conditional and not-conditional setups.
2019
MaL2CSec 2019 : Workshop on Machine Learning for Cyber-Crime Investigation and Cybersecurity
generative-adversarial-models; attacks-against-machine-learning; adversarial-input
04 Pubblicazione in atti di convegno::04b Atto di convegno in volume
Adversarial out-domain examples for generative models / Pasquini, Dario; Mingione, Marco; Bernaschi, Massimo. - (2019), pp. 272-280. (Intervento presentato al convegno MaL2CSec 2019 : Workshop on Machine Learning for Cyber-Crime Investigation and Cybersecurity tenutosi a Stockholm; Sweden) [10.1109/EuroSPW.2019.00037].
File allegati a questo prodotto
File Dimensione Formato  
Pasquini_Adversarial_2019.pdf

solo gestori archivio

Tipologia: Versione editoriale (versione pubblicata con il layout dell'editore)
Licenza: Tutti i diritti riservati (All rights reserved)
Dimensione 384.37 kB
Formato Adobe PDF
384.37 kB Adobe PDF   Contatta l'autore

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11573/1308016
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 7
  • ???jsp.display-item.citation.isi??? 6
social impact