TADA: Temporal Adversarial Data Augmentation for Time Series Data

TL;DR

This paper introduces TADA, a novel data augmentation method for time series that uses time warping and frequency domain techniques to improve model robustness against distribution shifts.

Contribution

TADA uniquely incorporates time warping into adversarial data augmentation for time series, addressing non-differentiability via frequency domain duality.

Findings

TADA outperforms existing domain generalization methods on various datasets.

Distribution visualization shows TADA induces realistic distribution shifts.

TADA effectively simulates real-world temporal distribution changes.

Abstract

Domain generalization aim to train models to effectively perform on samples that are unseen and outside of the distribution. Adversarial data augmentation (ADA) is a widely used technique in domain generalization. It enhances the model robustness by including synthetic samples designed to simulate potential unseen scenarios into the training datasets, which is then used to train the model. However, in time series data, traditional ADA approaches often fail to address distribution shifts related to temporal characteristics. To address this limitation, we propose Temporal Adversarial Data Augmentation (TADA) for time series data, which incorporate time warping into ADA. Although time warping is inherently non-differentiable, ADA relies on generating samples through backpropagation. We resolve this issue by leveraging the duality between phase shifts in the frequency domain and time shifts in the time domain, thereby making the process differentiable. Our evaluations across various time series datasets demonstrate that TADA outperforms existing methods for domain generalization. In addition, using distribution visualization, we confirmed that the distribution shifts induced by TADA are clearly different from those induced by ADA, and together, they effectively simulate real-world distribution shifts.