Learning to Infer Fast by Attending to Sparse Temporal Observations

Mehrnaz Motamed, Harry Bendekgey, Debora Sujono, Erik B. Sudderth
Proceedings of the 42nd Conference on Uncertainty in Artificial Intelligence, PMLR 337:4634-4660, 2026.

Abstract

Many scientific datasets contain sparse, irregular observations of dynamic processes. Fast inference techniques are required to infer the unseen evolution of physical systems underlying growing datasets, and estimate the parameters that govern them. We first accelerate the convergence of Gauss-{Markov} Variational Inference via efficient natural gradient computation and automatic step-size adaptation. We then propose two general-purpose schemes for initialization from sparse data, one using junction tree representations of Gauss-{Markov} distributions, and another attention-based neural network architecture that learns to predict correlated temporal posteriors. We show that these initializers dramatically improve performance for ecological models of soil decomposition and predator-prey dynamics. We further extend our methods to a semiparametric epidemiological model of SARS-CoV-2 RNA wastewater measurements, which uses {Gaussian} processes to model uncertainty in latent reproduction numbers. Our methods allow for quick inference of both local states and global dynamical parameters, and unlike prior work, can be applied even when measurements are sparse and the generative process is not fully known.

Cite this Paper

BibTeX
@InProceedings{pmlr-v337-motamed26a, title = {Learning to Infer Fast by Attending to Sparse Temporal Observations}, author = {Motamed, Mehrnaz and Bendekgey, Harry and Sujono, Debora and Sudderth, Erik B.}, booktitle = {Proceedings of the 42nd Conference on Uncertainty in Artificial Intelligence}, pages = {4634--4660}, year = {2026}, editor = {Perković, Emilija and Malinsky, Daniel}, volume = {337}, series = {Proceedings of Machine Learning Research}, month = {17--21 Aug}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v337/main/assets/motamed26a/motamed26a.pdf}, url = {https://proceedings.mlr.press/v337/motamed26a.html}, abstract = {Many scientific datasets contain sparse, irregular observations of dynamic processes. Fast inference techniques are required to infer the unseen evolution of physical systems underlying growing datasets, and estimate the parameters that govern them. We first accelerate the convergence of Gauss-{Markov} Variational Inference via efficient natural gradient computation and automatic step-size adaptation. We then propose two general-purpose schemes for initialization from sparse data, one using junction tree representations of Gauss-{Markov} distributions, and another attention-based neural network architecture that learns to predict correlated temporal posteriors. We show that these initializers dramatically improve performance for ecological models of soil decomposition and predator-prey dynamics. We further extend our methods to a semiparametric epidemiological model of SARS-CoV-2 RNA wastewater measurements, which uses {Gaussian} processes to model uncertainty in latent reproduction numbers. Our methods allow for quick inference of both local states and global dynamical parameters, and unlike prior work, can be applied even when measurements are sparse and the generative process is not fully known.} }
Endnote
%0 Conference Paper %T Learning to Infer Fast by Attending to Sparse Temporal Observations %A Mehrnaz Motamed %A Harry Bendekgey %A Debora Sujono %A Erik B. Sudderth %B Proceedings of the 42nd Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2026 %E Emilija Perković %E Daniel Malinsky %F pmlr-v337-motamed26a %I PMLR %P 4634--4660 %U https://proceedings.mlr.press/v337/motamed26a.html %V 337 %X Many scientific datasets contain sparse, irregular observations of dynamic processes. Fast inference techniques are required to infer the unseen evolution of physical systems underlying growing datasets, and estimate the parameters that govern them. We first accelerate the convergence of Gauss-{Markov} Variational Inference via efficient natural gradient computation and automatic step-size adaptation. We then propose two general-purpose schemes for initialization from sparse data, one using junction tree representations of Gauss-{Markov} distributions, and another attention-based neural network architecture that learns to predict correlated temporal posteriors. We show that these initializers dramatically improve performance for ecological models of soil decomposition and predator-prey dynamics. We further extend our methods to a semiparametric epidemiological model of SARS-CoV-2 RNA wastewater measurements, which uses {Gaussian} processes to model uncertainty in latent reproduction numbers. Our methods allow for quick inference of both local states and global dynamical parameters, and unlike prior work, can be applied even when measurements are sparse and the generative process is not fully known.
APA
Motamed, M., Bendekgey, H., Sujono, D. & Sudderth, E.B.. (2026). Learning to Infer Fast by Attending to Sparse Temporal Observations. Proceedings of the 42nd Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research 337:4634-4660 Available from https://proceedings.mlr.press/v337/motamed26a.html.

Related Material