Adversarial Examples as an Input-Fault Tolerance Problem

11/30/2018

∙

by   Angus Galloway, et al.

∙

0

∙

share

We analyze the adversarial examples problem in terms of a model's fault tolerance with respect to its input. Whereas previous work focuses on arbitrarily strict threat models, i.e., ϵ-perturbations, we consider arbitrary valid inputs and propose an information-based characteristic for evaluating tolerance to diverse input faults.