L-GTA: Latent Generative Modeling for Time Series Augmentation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Roque, Luis, Soares, Carlos, Cerqueira, Vitor, Torgo, Luis
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912513325531136
author Roque, Luis
Soares, Carlos
Cerqueira, Vitor
Torgo, Luis
author_facet Roque, Luis
Soares, Carlos
Cerqueira, Vitor
Torgo, Luis
contents Data augmentation is gaining importance across various aspects of time series analysis, from forecasting to classification and anomaly detection tasks. We introduce the Latent Generative Transformer Augmentation (L-GTA) model, a generative approach using a transformer-based variational recurrent autoencoder. This model uses controlled transformations within the latent space of the model to generate new time series that preserve the intrinsic properties of the original dataset. L-GTA enables the application of diverse transformations, ranging from simple jittering to magnitude warping, and combining these basic transformations to generate more complex synthetic time series datasets. Our evaluation of several real-world datasets demonstrates the ability of L-GTA to produce more reliable, consistent, and controllable augmented data. This translates into significant improvements in predictive accuracy and similarity measures compared to direct transformation methods.
format Preprint
id arxiv_https___arxiv_org_abs_2507_23615
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle L-GTA: Latent Generative Modeling for Time Series Augmentation
Roque, Luis
Soares, Carlos
Cerqueira, Vitor
Torgo, Luis
Machine Learning
Artificial Intelligence
68T01
I.5.1; G.3; H.2.8; I.2.1
Data augmentation is gaining importance across various aspects of time series analysis, from forecasting to classification and anomaly detection tasks. We introduce the Latent Generative Transformer Augmentation (L-GTA) model, a generative approach using a transformer-based variational recurrent autoencoder. This model uses controlled transformations within the latent space of the model to generate new time series that preserve the intrinsic properties of the original dataset. L-GTA enables the application of diverse transformations, ranging from simple jittering to magnitude warping, and combining these basic transformations to generate more complex synthetic time series datasets. Our evaluation of several real-world datasets demonstrates the ability of L-GTA to produce more reliable, consistent, and controllable augmented data. This translates into significant improvements in predictive accuracy and similarity measures compared to direct transformation methods.
title L-GTA: Latent Generative Modeling for Time Series Augmentation
topic Machine Learning
Artificial Intelligence
68T01
I.5.1; G.3; H.2.8; I.2.1
url https://arxiv.org/abs/2507.23615