Provably Robust Deep Learning via Adversarially Trained Smoothed Classifiers
This addresses the critical issue of adversarial vulnerability in neural networks for security-sensitive applications, representing an incremental but impactful advance in provable defense methods.
The paper tackles the problem of improving provable robustness in deep learning classifiers against adversarial perturbations by combining adversarial training with randomized smoothing, resulting in state-of-the-art performance on ImageNet and CIFAR-10 with significant gains.
Recent works have shown the effectiveness of randomized smoothing as a scalable technique for building neural network-based classifiers that are provably robust to $\ell_2$-norm adversarial perturbations. In this paper, we employ adversarial training to improve the performance of randomized smoothing. We design an adapted attack for smoothed classifiers, and we show how this attack can be used in an adversarial training setting to boost the provable robustness of smoothed classifiers. We demonstrate through extensive experimentation that our method consistently outperforms all existing provably $\ell_2$-robust classifiers by a significant margin on ImageNet and CIFAR-10, establishing the state-of-the-art for provable $\ell_2$-defenses. Moreover, we find that pre-training and semi-supervised learning boost adversarially trained smoothed classifiers even further. Our code and trained models are available at http://github.com/Hadisalman/smoothing-adversarial .