Timezone: »

Poster
Make Some Noise: Reliable and Efficient Single-Step Adversarial Training
Pau de Jorge Aranda · Adel Bibi · Riccardo Volpi · Amartya Sanyal · Philip Torr · Gregory Rogez · Puneet Dokania

Wed Nov 30 09:00 AM -- 11:00 AM (PST) @ Hall J #708
Recently, Wong et al. (2020) showed that adversarial training with single-step FGSM leads to a characteristic failure mode named catastrophic overfitting (CO), in which a model becomes suddenly vulnerable to multi-step attacks. Experimentally they showed that simply adding a random perturbation prior to FGSM (RS-FGSM) could prevent CO. However, Andriushchenko & Flammarion (2020) observed that RS-FGSM still leads to CO for larger perturbations, and proposed a computationally expensive regularizer (GradAlign) to avoid it. In this work, we methodically revisit the role of noise and clipping in single-step adversarial training. Contrary to previous intuitions, we find that using a stronger noise around the clean sample combined with \textit{not clipping} is highly effective in avoiding CO for large perturbation radii. We then propose Noise-FGSM (N-FGSM) that, while providing the benefits of single-step adversarial training, does not suffer from CO. Empirical analyses on a large suite of experiments show that N-FGSM is able to match or surpass the performance of previous state of-the-art GradAlign while achieving 3$\times$ speed-up.

#### Author Information

##### Pau de Jorge Aranda (University of Oxford &amp; Naver Labs Europe)

I'm a a PhD student at the University of Oxford and Naver Labs Europe. My research interests include but are not limited to deep learning, computer vision, and machine learning.