Empirical Study of Mix-based Data Augmentation Methods in Physiological Time Series Data

TL;DR

This paper systematically evaluates mix-based data augmentation methods like mixup, cutmix, and manifold mixup on physiological time series datasets, showing consistent performance improvements without extensive tuning.

Contribution

It is the first comprehensive study applying mix-based augmentations to physiological time series classification, demonstrating their effectiveness and ease of use.

Findings

Mix-based augmentations improve classification accuracy across datasets.

Performance gains are achieved without expert knowledge or extensive tuning.

The methods have unique properties beneficial for physiological data.

Abstract

Data augmentation is a common practice to help generalization in the procedure of deep model training. In the context of physiological time series classification, previous research has primarily focused on label-invariant data augmentation methods. However, another class of augmentation techniques (\textit{i.e., Mixup}) that emerged in the computer vision field has yet to be fully explored in the time series domain. In this study, we systematically review the mix-based augmentations, including mixup, cutmix, and manifold mixup, on six physiological datasets, evaluating their performance across different sensory data and classification tasks. Our results demonstrate that the three mix-based augmentations can consistently improve the performance on the six datasets. More importantly, the improvement does not rely on expert knowledge or extensive parameter tuning. Lastly, we provide an overview of the unique properties of the mix-based augmentation methods and highlight the potential benefits of using the mix-based augmentation in physiological time series data.