Rethinking Image-Scaling Attacks: The Interplay Between Vulnerabilities in Machine Learning Systems
Yue Gao, Ilia Shumailov, Kassem Fawaz
Abstract
As real-world images come in varying sizes, the machine learning model is part of a larger system that includes an upstream image scaling algorithm. In this paper, we investigate the interplay between vulnerabilities of the image scaling procedure and machine learning models in the decision-based black-box setting. We propose a novel sampling strategy to make a black-box attack exploit vulnerabilities in scaling algorithms, scaling defenses, and the final machine learning model in an end-to-end manner. Based on this scaling-aware attack, we reveal that most existing scaling defenses are ineffective under threat from downstream models. Moreover, we empirically observe that standard black-box attacks can significantly improve their performance by exploiting the vulnerable scaling procedure. We further demonstrate this problem on a commercial Image Analysis API with decision-based black-box attacks.
BibTeX
@InProceedings{pmlr-v162-gao22g,
title = {Rethinking Image-Scaling Attacks: The Interplay Between Vulnerabilities in Machine Learning Systems},
author = {Gao, Yue and Shumailov, Ilia and Fawaz, Kassem},
booktitle = {Proceedings of the 39th International Conference on Machine Learning},
pages = {7102--7121},
year = {2022},
editor = {Chaudhuri, Kamalika and Jegelka, Stefanie and Song, Le and Szepesvari, Csaba and Niu, Gang and Sabato, Sivan},
volume = {162},
series = {Proceedings of Machine Learning Research},
month = {17--23 Jul},
publisher = {PMLR},
pdf = {https://proceedings.mlr.press/v162/gao22g/gao22g.pdf},
url = {https://proceedings.mlr.press/v162/gao22g.html},
abstract = {As real-world images come in varying sizes, the machine learning model is part of a larger system that includes an upstream image scaling algorithm. In this paper, we investigate the interplay between vulnerabilities of the image scaling procedure and machine learning models in the decision-based black-box setting. We propose a novel sampling strategy to make a black-box attack exploit vulnerabilities in scaling algorithms, scaling defenses, and the final machine learning model in an end-to-end manner. Based on this scaling-aware attack, we reveal that most existing scaling defenses are ineffective under threat from downstream models. Moreover, we empirically observe that standard black-box attacks can significantly improve their performance by exploiting the vulnerable scaling procedure. We further demonstrate this problem on a commercial Image Analysis API with decision-based black-box attacks.}
}