Improving the Adversarial Robustness and Interpretability of Deep Neural Networks by Regularizing their Input Gradients
Andrew Slavin Ross
arXiv e-Print archive - 2017 via Local arXiv
cs.LG, cs.CR, cs.CV
First published: 2017/11/26 (4 years ago) Abstract: Deep neural networks have proven remarkably effective at solving many
classification problems, but have been criticized recently for two major
weaknesses: the reasons behind their predictions are uninterpretable, and the
predictions themselves can often be fooled by small adversarial perturbations.
These problems pose major obstacles for the adoption of neural networks in
domains that require security or transparency. In this work, we evaluate the
effectiveness of defenses that differentiably penalize the degree to which
small changes in inputs can alter model predictions. Across multiple attacks,
architectures, defenses, and datasets, we find that neural networks trained
with this input gradient regularization exhibit robustness to transferred
adversarial examples generated to fool all of the other models. We also find
that adversarial examples generated to fool gradient-regularized models fool
all other models equally well, and actually lead to more "legitimate,"
interpretable misclassifications as rated by people (which we confirm in a
human subject experiment). Finally, we demonstrate that regularizing input
gradients makes them more naturally interpretable as rationales for model
predictions. We conclude by discussing this relationship between
interpretability and robustness in deep neural networks.