DiffStyleTS: Diffusion Model for Style Transfer in Time Series

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Nagda, Mayank, Ostheimer, Phil, Arweiler, Justus, Jungjohann, Indra, Werner, Jennifer, Wagner, Dennis, Muraleedharan, Aparna, Jafari, Pouya, Schmid, Jochen, Jirasek, Fabian, Burger, Jakob, Bortz, Michael, Hasse, Hans, Mandt, Stephan, Kloft, Marius, Fellenz, Sophie
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914089870032896
author Nagda, Mayank
Ostheimer, Phil
Arweiler, Justus
Jungjohann, Indra
Werner, Jennifer
Wagner, Dennis
Muraleedharan, Aparna
Jafari, Pouya
Schmid, Jochen
Jirasek, Fabian
Burger, Jakob
Bortz, Michael
Hasse, Hans
Mandt, Stephan
Kloft, Marius
Fellenz, Sophie
author_facet Nagda, Mayank
Ostheimer, Phil
Arweiler, Justus
Jungjohann, Indra
Werner, Jennifer
Wagner, Dennis
Muraleedharan, Aparna
Jafari, Pouya
Schmid, Jochen
Jirasek, Fabian
Burger, Jakob
Bortz, Michael
Hasse, Hans
Mandt, Stephan
Kloft, Marius
Fellenz, Sophie
contents Style transfer combines the content of one signal with the style of another. It supports applications such as data augmentation and scenario simulation, helping machine learning models generalize in data-scarce domains. While well developed in vision and language, style transfer methods for time series data remain limited. We introduce DiffTSST, a diffusion-based framework that disentangles a time series into content and style representations via convolutional encoders and recombines them through a self-supervised attention-based diffusion process. At inference, encoders extract content and style from two distinct series, enabling conditional generation of novel samples to achieve style transfer. We demonstrate both qualitatively and quantitatively that DiffTSST achieves effective style transfer. We further validate its real-world utility by showing that data augmentation with DiffTSST improves anomaly detection in data-scarce regimes.
format Preprint
id arxiv_https___arxiv_org_abs_2510_11335
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle DiffStyleTS: Diffusion Model for Style Transfer in Time Series
Nagda, Mayank
Ostheimer, Phil
Arweiler, Justus
Jungjohann, Indra
Werner, Jennifer
Wagner, Dennis
Muraleedharan, Aparna
Jafari, Pouya
Schmid, Jochen
Jirasek, Fabian
Burger, Jakob
Bortz, Michael
Hasse, Hans
Mandt, Stephan
Kloft, Marius
Fellenz, Sophie
Machine Learning
Style transfer combines the content of one signal with the style of another. It supports applications such as data augmentation and scenario simulation, helping machine learning models generalize in data-scarce domains. While well developed in vision and language, style transfer methods for time series data remain limited. We introduce DiffTSST, a diffusion-based framework that disentangles a time series into content and style representations via convolutional encoders and recombines them through a self-supervised attention-based diffusion process. At inference, encoders extract content and style from two distinct series, enabling conditional generation of novel samples to achieve style transfer. We demonstrate both qualitatively and quantitatively that DiffTSST achieves effective style transfer. We further validate its real-world utility by showing that data augmentation with DiffTSST improves anomaly detection in data-scarce regimes.
title DiffStyleTS: Diffusion Model for Style Transfer in Time Series
topic Machine Learning
url https://arxiv.org/abs/2510.11335