An Explainable Adversarial Robustness Metric for Deep Learning Neural Networks

TL;DR

This paper introduces Noise Sensitivity Score (NSS), a novel explainable metric for evaluating deep neural networks' robustness against adversarial attacks, validated across multiple datasets and architectures.

Contribution

The paper proposes NSS, an explainable and mathematically grounded robustness metric for DNNs, addressing the gap in existing evaluation methods for adversarial performance.

Findings

NSS effectively quantifies DNN robustness on specific inputs.

A dataset-level robustness metric based on skewness is introduced.

Extensive experiments validate the generalization of the proposed metrics.

Abstract

Deep Neural Networks(DNN) have excessively advanced the field of computer vision by achieving state of the art performance in various vision tasks. These results are not limited to the field of vision but can also be seen in speech recognition and machine translation tasks. Recently, DNNs are found to poorly fail when tested with samples that are crafted by making imperceptible changes to the original input images. This causes a gap between the validation and adversarial performance of a DNN. An effective and generalizable robustness metric for evaluating the performance of DNN on these adversarial inputs is still missing from the literature. In this paper, we propose Noise Sensitivity Score (NSS), a metric that quantifies the performance of a DNN on a specific input under different forms of fix-directional attacks. An insightful mathematical explanation is provided for deeply understanding the proposed metric. By leveraging the NSS, we also proposed a skewness based dataset robustness metric for evaluating a DNN's adversarial performance on a given dataset. Extensive experiments using widely used state of the art architectures along with popular classification datasets, such as MNIST, CIFAR-10, CIFAR-100, and ImageNet, are used to validate the effectiveness and generalization of our proposed metrics. Instead of simply measuring a DNN's adversarial robustness in the input domain, as previous works, the proposed NSS is built on top of insightful mathematical understanding of the adversarial attack and gives a more explicit explanation of the robustness.