Adversarial Robustness via Robust Low Rank Representations

Abstract

Adversarial robustness measures the susceptibility of a classifier to imperceptible perturbations made to the inputs at test time. In this work we highlight the benefits of natural low rank representations that often exist for real data such as images, for training neural networks with certified robustness guarantees.

PDF NeurIPS Semantic Scholar

Cite

Text

Awasthi et al. "Adversarial Robustness via Robust Low Rank Representations." Neural Information Processing Systems, 2020.

Markdown

[Awasthi et al. "Adversarial Robustness via Robust Low Rank Representations." Neural Information Processing Systems, 2020.](https://mlanthology.org/neurips/2020/awasthi2020neurips-adversarial/)

BibTeX

@inproceedings{awasthi2020neurips-adversarial,
  title     = {{Adversarial Robustness via Robust Low Rank Representations}},
  author    = {Awasthi, Pranjal and Jain, Himanshu and Rawat, Ankit Singh and Vijayaraghavan, Aravindan},
  booktitle = {Neural Information Processing Systems},
  year      = {2020},
  url       = {https://mlanthology.org/neurips/2020/awasthi2020neurips-adversarial/}
}