Lipschitz Networks and Distributional Robustness
Machine Learning
2018-09-06 v1 Machine Learning
Abstract
Robust risk minimisation has several advantages: it has been studied with regards to improving the generalisation properties of models and robustness to adversarial perturbation. We bound the distributionally robust risk for a model class rich enough to include deep neural networks by a regularised empirical risk involving the Lipschitz constant of the model. This allows us to interpretand quantify the robustness properties of a deep neural network. As an application we show the distributionally robust risk upperbounds the adversarial training risk.
Cite
@article{arxiv.1809.01129,
title = {Lipschitz Networks and Distributional Robustness},
author = {Zac Cranko and Simon Kornblith and Zhan Shi and Richard Nock},
journal= {arXiv preprint arXiv:1809.01129},
year = {2018}
}