AISTATS 2023poster37 citations

Theoretically Grounded Loss Functions and Algorithms for Adversarial Robustness

Pranjal Awasthi, Anqi Mao, Mehryar Mohri, Yutao Zhong

Abstract

Adversarial robustness is a critical property of classifiers in applications as they are increasingly deployed in complex real-world systems. Yet, achieving accurate adversarial robustness in machine learning remains a persistent challenge and the choice of the surrogate loss function used for training a key factor. We present a family of new loss functions for adversarial robustness, smooth adversarial losses, which we show can be derived in a general way from broad families of loss functions used in multi-class classification. We prove strong H-consistency theoretical guarantees for these loss functions, including multi-class H-consistency bounds for sum losses in the adversarial setting. We design new regularized algorithms based on the minimization of these principled smooth adversarial losses (PSAL). We further show through a series of extensive experiments with the CIFAR-10, CIFAR-100 and SVHN datasets that our PSAL algorithm consistently outperforms the current state-of-the-art technique, TRADES, for both robust accuracy against l-infinity-norm bounded perturbations and, even more significantly, for clean accuracy. Finally, we prove that, unlike PSAL, the TRADES loss in general does not admit an H-consistency property.

BibTeX
@InProceedings{pmlr-v206-awasthi23c,
  title = 	 {Theoretically Grounded Loss Functions and Algorithms for Adversarial Robustness},
  author =       {Awasthi, Pranjal and Mao, Anqi and Mohri, Mehryar and Zhong, Yutao},
  booktitle = 	 {Proceedings of The 26th International Conference on Artificial Intelligence and Statistics},
  pages = 	 {10077--10094},
  year = 	 {2023},
  editor = 	 {Ruiz, Francisco and Dy, Jennifer and van de Meent, Jan-Willem},
  volume = 	 {206},
  series = 	 {Proceedings of Machine Learning Research},
  month = 	 {25--27 Apr},
  publisher =    {PMLR},
  pdf = 	 {https://proceedings.mlr.press/v206/awasthi23c/awasthi23c.pdf},
  url = 	 {https://proceedings.mlr.press/v206/awasthi23c.html},
  abstract = 	 {Adversarial robustness is a critical property of classifiers in applications as they are increasingly deployed in complex real-world systems. Yet, achieving accurate adversarial robustness in machine learning remains a persistent challenge and the choice of the surrogate loss function used for training a key factor. We present a family of new loss functions for adversarial robustness, smooth adversarial losses, which we show can be derived in a general way from broad families of loss functions used in multi-class classification. We prove strong H-consistency theoretical guarantees for these loss functions, including multi-class H-consistency bounds for sum losses in the adversarial setting. We design new regularized algorithms based on the minimization of these principled smooth adversarial losses (PSAL). We further show through a series of extensive experiments with the CIFAR-10, CIFAR-100 and SVHN datasets that our PSAL algorithm consistently outperforms the current state-of-the-art technique, TRADES, for both robust accuracy against l-infinity-norm bounded perturbations and, even more significantly, for clean accuracy. Finally, we prove that, unlike PSAL, the TRADES loss in general does not admit an H-consistency property.}
}