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.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.