D3A-TS: Denoising-Driven Data Augmentation in Time Series
This work addresses data scarcity in time series domains such as engineering, offering a synthetic data generation approach, though it is incremental as it adapts existing diffusion models from image processing.
The paper tackled the problem of scarce data in time series tasks like predictive maintenance by proposing a diffusion model-based data augmentation method, demonstrating its utility in improving classification and regression models across six diverse datasets.
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.