Timezone: »

 
Poster
Reliable evaluation of adversarial robustness with an ensemble of diverse parameter-free attacks
Francesco Croce · Matthias Hein

Wed Jul 15 02:00 PM -- 02:45 PM & Thu Jul 16 01:00 AM -- 01:45 AM (PDT) @ None #None
in Poster Session 28 »

The field of defense strategies against adversarial attacks has significantly grown over the last years, but progress is hampered as the evaluation of adversarial defenses is often insufficient and thus gives a wrong impression of robustness. Many promising defenses could be broken later on, making it difficult to identify the state-of-the-art. Frequent pitfalls in the evaluation are improper tuning of hyperparameters of the attacks, gradient obfuscation or masking. In this paper we first propose two extensions of the PGD-attack overcoming failures due to suboptimal step size and problems of the objective function. We then combine our novel attacks with two complementary existing ones to form a parameter-free, computationally affordable and user-independent ensemble of attacks to test adversarial robustness. We apply our ensemble to over 50 models from papers published at recent top machine learning and computer vision venues. In all except one of the cases we achieve lower robust test accuracy than reported in these papers, often by more than 10\%, identifying several broken defenses.

Keywords: Adversarial Examples

Author Information

Francesco Croce (University of Tuebingen)
Matthias Hein (University of Tübingen)

More from the Same Authors