D3A-TS: Denoising-Driven Data Augmentation in Time Series

TL;DR

This paper introduces D3A-TS, a novel data augmentation method for time series using diffusion probabilistic models and meta-attributes, significantly improving model training in data-scarce scenarios.

Contribution

It pioneers the application of diffusion models for time series data augmentation, incorporating meta-attributes to enhance synthetic data quality and model performance.

Findings

Effective augmentation across diverse datasets

Improved classification and regression accuracy

Validated through extensive ablation studies

Abstract

It has been demonstrated that the amount of data is crucial in data-driven machine learning methods. Data is always valuable, but in some tasks, it is almost like gold. This occurs in engineering areas where data is scarce or very expensive to obtain, such as predictive maintenance, where faults are rare. In this context, a mechanism to generate synthetic data can be very useful. While in fields such as Computer Vision or Natural Language Processing synthetic data generation has been extensively explored with promising results, in other domains such as time series it has received less attention. This work specifically focuses on studying and analyzing the use of different techniques for data augmentation in time series for classification and regression problems. The proposed approach involves the use of diffusion probabilistic models, which have recently achieved successful results in the field of Image Processing, for data augmentation in time series. Additionally, the use of meta-attributes to condition the data augmentation process is investigated. The results highlight the high utility of this methodology in creating synthetic data to train classification and regression models. To assess the results, six different datasets from diverse domains were employed, showcasing versatility in terms of input size and output types. Finally, an extensive ablation study is conducted to further support the obtained outcomes.