LSCD: Lomb–Scargle Conditioned Diffusion for Time series Imputation

Elizabeth Fons, Alejandro Sztrajman, Yousef El-Laham, Luciana Ferrer, Svitlana Vyetrenko, Manuela Veloso
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:17411-17436, 2025.

Abstract

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.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-fons25a, title = {{LSCD}: Lomb–Scargle Conditioned Diffusion for Time series Imputation}, author = {Fons, Elizabeth and Sztrajman, Alejandro and El-Laham, Yousef and Ferrer, Luciana and Vyetrenko, Svitlana and Veloso, Manuela}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {17411--17436}, year = {2025}, editor = {Singh, Aarti and Fazel, Maryam and Hsu, Daniel and Lacoste-Julien, Simon and Berkenkamp, Felix and Maharaj, Tegan and Wagstaff, Kiri and Zhu, Jerry}, volume = {267}, series = {Proceedings of Machine Learning Research}, month = {13--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v267/main/assets/fons25a/fons25a.pdf}, url = {https://proceedings.mlr.press/v267/fons25a.html}, abstract = {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.} }
Endnote
%0 Conference Paper %T LSCD: Lomb–Scargle Conditioned Diffusion for Time series Imputation %A Elizabeth Fons %A Alejandro Sztrajman %A Yousef El-Laham %A Luciana Ferrer %A Svitlana Vyetrenko %A Manuela Veloso %B Proceedings of the 42nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Aarti Singh %E Maryam Fazel %E Daniel Hsu %E Simon Lacoste-Julien %E Felix Berkenkamp %E Tegan Maharaj %E Kiri Wagstaff %E Jerry Zhu %F pmlr-v267-fons25a %I PMLR %P 17411--17436 %U https://proceedings.mlr.press/v267/fons25a.html %V 267 %X 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.
APA
Fons, E., Sztrajman, A., El-Laham, Y., Ferrer, L., Vyetrenko, S. & Veloso, M.. (2025). LSCD: Lomb–Scargle Conditioned Diffusion for Time series Imputation. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:17411-17436 Available from https://proceedings.mlr.press/v267/fons25a.html.

Related Material