Robustly Learning a Single Neuron via Sharpness

Abstract

We study the problem of learning a single neuron with respect to the $L_2^2$-loss in the presence of adversarial label noise. We give an efficient algorithm that, for a broad family of activations including ReLUs, approximates the optimal $L_2^2$-error within a constant factor. Notably, our algorithm succeeds under much milder distributional assumptions compared to prior work. The key ingredient enabling our results is a novel connection to local error bounds from optimization theory.

Cite

Text

Wang et al. "Robustly Learning a Single Neuron via Sharpness." International Conference on Machine Learning, 2023.

Markdown

[Wang et al. "Robustly Learning a Single Neuron via Sharpness." International Conference on Machine Learning, 2023.](https://mlanthology.org/icml/2023/wang2023icml-robustly/)

BibTeX

@inproceedings{wang2023icml-robustly,
  title     = {{Robustly Learning a Single Neuron via Sharpness}},
  author    = {Wang, Puqian and Zarifis, Nikos and Diakonikolas, Ilias and Diakonikolas, Jelena},
  booktitle = {International Conference on Machine Learning},
  year      = {2023},
  pages     = {36541-36577},
  volume    = {202},
  url       = {https://mlanthology.org/icml/2023/wang2023icml-robustly/}
}