Approximate inference in latent Gaussian-Markov models from continuous time observations

Botond Cseke, Manfred Opper, Guido Sanguinetti

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

We propose an approximate inference algorithm for continuous time Gaussian-Markov process models with both discrete and continuous time likelihoods. We show that the continuous time limit of the expectation propagation algorithm exists and results in a hybrid fixed point iteration consisting of (1) expectation propagation updates for the discrete time terms and (2) variational updates for the continuous time term. We introduce corrections methods that improve on the marginals of the approximation. This approach extends the classical Kalman-Bucy smoothing procedure to non-Gaussian observations, enabling continuous-time inference in a variety of models, including spiking neuronal models (state-space models with point process observations) and box likelihood models. Experimental results on real and simulated data demonstrate high distributional accuracy and significant computational savings compared to discrete-time approaches in a neural application.
Original languageEnglish
Title of host publicationAdvances in Neural Information Processing Systems 26
PublisherCurran Associates Inc
Pages971
Number of pages979
Publication statusPublished - 2013

Fingerprint

Dive into the research topics of 'Approximate inference in latent Gaussian-Markov models from continuous time observations'. Together they form a unique fingerprint.

Cite this