Dynamic Time Warping based Adversarial Framework for Time-Series Domain

TL;DR

This paper introduces DTW-AR, a novel adversarial framework for time-series data that leverages dynamic time warping to generate diverse adversarial examples and enhance neural network robustness.

Contribution

It proposes a new DTW-based adversarial framework for time-series data, with a theoretically justified algorithm for creating diverse adversarial examples.

Findings

DTW-AR effectively fools DNNs on time-series benchmarks.

Adversarial training with DTW-AR improves model robustness.

DTW outperforms Euclidean distance in adversarial example generation.

Abstract

Despite the rapid progress on research in adversarial robustness of deep neural networks (DNNs), there is little principled work for the time-series domain. Since time-series data arises in diverse applications including mobile health, finance, and smart grid, it is important to verify and improve the robustness of DNNs for the time-series domain. In this paper, we propose a novel framework for the time-series domain referred as {\em Dynamic Time Warping for Adversarial Robustness (DTW-AR)} using the dynamic time warping measure. Theoretical and empirical evidence is provided to demonstrate the effectiveness of DTW over the standard Euclidean distance metric employed in prior methods for the image domain. We develop a principled algorithm justified by theoretical analysis to efficiently create diverse adversarial examples using random alignment paths. Experiments on diverse real-world benchmarks show the effectiveness of DTW-AR to fool DNNs for time-series data and to improve their robustness using adversarial training. The source code of DTW-AR algorithms is available at https://github.com/tahabelkhouja/DTW-AR