Benutzer: Gast  Login
Titel:

Guessing Smart: Biased Sampling for Efficient Black-Box Adversarial Attacks

Dokumenttyp:
Konferenzbeitrag
Art des Konferenzbeitrags:
Textbeitrag / Aufsatz
Autor(en):
Brunner, Thomas; Diehl, Frederik; Truong Le, Michael; Knoll, Alois
Seitenangaben Beitrag:
4957-4965
Abstract:
We consider adversarial examples for image classification in the black-box decision-based setting. Here, an attacker cannot access confidence scores, but only the final label. Most attacks for this scenario are either unreliable or inefficient. Focusing on the latter, we show that a specific class of attacks, Boundary Attacks, can be reinterpreted as a biased sampling framework that gains efficiency from domain knowledge. We identify three such biases, image frequency, regional masks and surroga...     »
Kongress- / Buchtitel:
2019 IEEE/CVF International Conference on Computer Vision (ICCV)
Kongress / Zusatzinformationen:
Seoul, South Korea
Verlag / Institution:
IEEE
Jahr:
2019
Monat:
Oct
E-ISBN:
978-1-7281-4803-8
Reviewed:
ja
Volltext / DOI:
doi:10.1109/ICCV.2019.00506
Copyright Informationen:
© 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
 BibTeX