Robust Deep Neural Networks Inspired by Fuzzy Logic

TL;DR

This paper introduces fuzzy logic-inspired deep neural network architectures that enhance robustness against adversarial attacks and noise by promoting more localized decision regions, addressing flaws in traditional models.

Contribution

The paper proposes novel neural network architectures inspired by fuzzy logic, improving robustness and noise rejection compared to standard models.

Findings

Models are more local and reject noise samples effectively.

Enhanced robustness against adversarial examples demonstrated.

Ablation studies support the role of logic-inspired traits in robustness.

Abstract

Deep neural networks have achieved impressive performance and become the de-facto standard in many tasks. However, troubling phenomena such as adversarial and fooling examples suggest that the generalization they make is flawed. I argue that among the roots of the phenomena are two geometric properties of common deep learning architectures: their distributed nature and the connectedness of their decision regions. As a remedy, I propose new architectures inspired by fuzzy logic that combine several alternative design elements. Through experiments on MNIST and CIFAR-10, the new models are shown to be more local, better at rejecting noise samples, and more robust against adversarial examples. Ablation analyses reveal behaviors on adversarial examples that cannot be explained by the linearity hypothesis but are consistent with the hypothesis that logic-inspired traits create more robust models.