LSCD: Lomb-Scargle Conditioned Diffusion for Time series Imputation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Fons, Elizabeth, Sztrajman, Alejandro, El-Laham, Yousef, Ferrer, Luciana, Vyetrenko, Svitlana, Veloso, Manuela
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918066310348800
author Fons, Elizabeth
Sztrajman, Alejandro
El-Laham, Yousef
Ferrer, Luciana
Vyetrenko, Svitlana
Veloso, Manuela
author_facet Fons, Elizabeth
Sztrajman, Alejandro
El-Laham, Yousef
Ferrer, Luciana
Vyetrenko, Svitlana
Veloso, Manuela
contents Time series with missing or irregularly sampled data are a persistent challenge in machine learning. Many methods operate on the frequency-domain, relying on the Fast Fourier Transform (FFT) which assumes uniform sampling, therefore requiring prior interpolation that can distort the spectra. To address this limitation, we introduce a differentiable Lomb--Scargle layer that enables a reliable computation of the power spectrum of irregularly sampled data. We integrate this layer into a novel score-based diffusion model (LSCD) for time series imputation conditioned on the entire signal spectrum. Experiments on synthetic and real-world benchmarks demonstrate that our method recovers missing data more accurately than purely time-domain baselines, while simultaneously producing consistent frequency estimates. Crucially, our method can be easily integrated into learning frameworks, enabling broader adoption of spectral guidance in machine learning approaches involving incomplete or irregular data.
format Preprint
id arxiv_https___arxiv_org_abs_2506_17039
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle LSCD: Lomb-Scargle Conditioned Diffusion for Time series Imputation
Fons, Elizabeth
Sztrajman, Alejandro
El-Laham, Yousef
Ferrer, Luciana
Vyetrenko, Svitlana
Veloso, Manuela
Machine Learning
Artificial Intelligence
Time series with missing or irregularly sampled data are a persistent challenge in machine learning. Many methods operate on the frequency-domain, relying on the Fast Fourier Transform (FFT) which assumes uniform sampling, therefore requiring prior interpolation that can distort the spectra. To address this limitation, we introduce a differentiable Lomb--Scargle layer that enables a reliable computation of the power spectrum of irregularly sampled data. We integrate this layer into a novel score-based diffusion model (LSCD) for time series imputation conditioned on the entire signal spectrum. Experiments on synthetic and real-world benchmarks demonstrate that our method recovers missing data more accurately than purely time-domain baselines, while simultaneously producing consistent frequency estimates. Crucially, our method can be easily integrated into learning frameworks, enabling broader adoption of spectral guidance in machine learning approaches involving incomplete or irregular data.
title LSCD: Lomb-Scargle Conditioned Diffusion for Time series Imputation
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2506.17039