A single gradient step finds adversarial examples on random two-layers neural networks

Autor:	Bubeck, Sébastien, Cherapanamjeri, Yeshwanth, Gidel, Gauthier, Combes, Rémi Tachet des
Rok vydání:	2021
Předmět:	Computer Science - Machine Learning Computer Science - Cryptography and Security Statistics - Machine Learning
Druh dokumentu:	Working Paper
Popis:	Daniely and Schacham recently showed that gradient descent finds adversarial examples on random undercomplete two-layers ReLU neural networks. The term "undercomplete" refers to the fact that their proof only holds when the number of neurons is a vanishing fraction of the ambient dimension. We extend their result to the overcomplete case, where the number of neurons is larger than the dimension (yet also subexponential in the dimension). In fact we prove that a single step of gradient descent suffices. We also show this result for any subexponential width random neural network with smooth activation function. Comment: Added a comment about universal adversarial perturbations. 18 pages, 7 figures
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/2104.03863 Zobrazit plný text záznamu View this record from Arxiv