Paper published in a book (Scientific congresses, symposiums and conference proceedings)
LGV: Boosting Adversarial Example Transferability from Large Geometric Vicinity
Gubri, Martin; Cordy, Maxime; Papadakis, Mike et al.
2022In Computer Vision -- ECCV 2022
Peer reviewed
 

Files


Full Text
2207.13129.pdf
Author preprint (1.81 MB)
Preprint arXiv
Download

All documents in ORBilu are protected by a user license.

Send to



Details



Keywords :
adversarial machine learning; adversarial examples; deep learning; Transferability; Loss Geometry
Abstract :
[en] We propose transferability from Large Geometric Vicinity (LGV), a new technique to increase the transferability of black-box adversarial attacks. LGV starts from a pretrained surrogate model and collects multiple weight sets from a few additional training epochs with a constant and high learning rate. LGV exploits two geometric properties that we relate to transferability. First, models that belong to a wider weight optimum are better surrogates. Second, we identify a subspace able to generate an effective surrogate ensemble among this wider optimum. Through extensive experiments, we show that LGV alone outperforms all (combinations of) four established test-time transformations by 1.8 to 59.9\% points. Our findings shed new light on the importance of the geometry of the weight space to explain the transferability of adversarial examples.
Disciplines :
Computer science
Author, co-author :
Gubri, Martin ;  University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT) > SerVal
Cordy, Maxime  ;  University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT) > SerVal
Papadakis, Mike ;  University of Luxembourg > Faculty of Science, Technology and Medicine (FSTM) > Department of Computer Science (DCS)
Traon, Yves Le
Sen, Koushik
External co-authors :
yes
Language :
English
Title :
LGV: Boosting Adversarial Example Transferability from Large Geometric Vicinity
Publication date :
2022
Event name :
EUROPEAN CONFERENCE ON COMPUTER VISION
Event date :
October 23–27, 2022
Audience :
International
Main work title :
Computer Vision -- ECCV 2022
Publisher :
Springer Nature Switzerland
ISBN/EAN :
978-3-031-19772-7
Pages :
603--618
Peer reviewed :
Peer reviewed
Focus Area :
Security, Reliability and Trust
FnR Project :
FNR12669767 - Testing Self-learning Systems, 2018 (01/09/2019-31/08/2022) - Yves Le Traon
Available on ORBilu :
since 30 December 2022

Statistics


Number of views
69 (3 by Unilu)
Number of downloads
16 (1 by Unilu)

Scopus citations®
 
9
Scopus citations®
without self-citations
9

Bibliography


Similar publications



Contact ORBilu