Adversarial Examples as an Input-Fault Tolerance Problem

Abstract

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., $\epsilon$-perturbations, we consider arbitrary valid inputs and propose an information-based characteristic for evaluating tolerance to diverse input faults.

arXiv commentNIPS 2018 Workshop on Security and Machine Learning. Source available at https://github.com/uoguelph-mlrg/nips18-secml-advex-input-fault

References in corpus (13)

• Improved Regularization of Convolutional Neural Networks with Cutout
• The information bottleneck method
• Obfuscated Gradients Give a False Sense of Security: Circumventing Defenses to Adversarial Examples
• Adversarially Robust Generalization Requires More Data
• Technical Report on the CleverHans v2.1.0 Adversarial Examples Library
• Spatially Transformed Adversarial Examples
• Towards the first adversarially robust neural network model on MNIST
• Motivating the Rules of the Game for Adversarial Example Research
• Do CIFAR-10 Classifiers Generalize to CIFAR-10?
• Constructing Unrestricted Adversarial Examples with Generative Models
• Attacking the Madry Defense Model with $L_1$-based Adversarial Examples
• Unrestricted Adversarial Examples
• Adversarial Training Versus Weight Decay