Modeling Irregular Astronomical Time Series with Neural Stochastic Delay Differential Equations

Abstract

Astronomical time series from large-scale surveys like LSST are often irregularly sampled and incomplete, posing challenges for classification and anomaly detection. We introduce a new framework based on Neural Stochastic Delay Differential Equations (Neural SDDEs) that combines stochastic modeling with neural networks to capture delayed temporal dynamics and handle irregular observations. Our approach integrates a delay-aware neural architecture, a numerical solver for SDDEs, and mechanisms to robustly learn from noisy, sparse sequences. Experiments on irregularly sampled astronomical data demonstrate strong classification accuracy and effective detection of novel astrophysical events, even with partial labels. This work highlights Neural SDDEs as a principled and practical tool for time series analysis under observational constraints.

CIKM '25: Proceedings of the 34th ACM International Conference on Information and Knowledge Management. https://doi.org/10.1145/3746252.3760805

References in corpus (17)

• Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling
• LSST: from Science Drivers to Reference Design and Anticipated Data Products
• Deep learning for time series classification: a review
• The Zwicky Transient Facility: System Overview, Performance, and First Results
• A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networks
• Deep learning-based numerical methods for high-dimensional parabolic partial differential equations and backward stochastic differential equations
• A Description of Quasar Variability Measured Using Repeated SDSS and POSS Imaging
• Periodograms for Multiband Astronomical Time Series
• RAPID: Early Classification of Explosive Transients using Deep Learning
• Models and Simulations for the Photometric LSST Astronomical Time Series Classification Challenge (PLAsTiCC)
• The Zwicky Transient Facility Alert Distribution System
• A recurrent neural network for classification of unevenly sampled variable stars
• The Large Synoptic Survey Telescope as a Near-Earth Object Discovery Machine
• Learning Long-Term Dependencies in Irregularly-Sampled Time Series
• Using Host Galaxy Photometric Redshifts to Improve Cosmological Constraints with Type Ia Supernova in the LSST Era
• Latent Stochastic Differential Equations for Modeling Quasar Variability and Inferring Black Hole Properties
• Delay-SDE-net: A deep learning approach for time series modelling with memory and uncertainty estimates