Inference in Latent Force Models Using Optimal State Estimation

TL;DR

This paper introduces two optimal state estimation methods for latent force models that incorporate system constraints, demonstrated through numerical and real-world biomedical examples, improving force reconstruction accuracy.

Contribution

The paper presents novel optimal state estimators for latent force models that explicitly consider system constraints, advancing the state-of-the-art in force and state estimation.

Findings

Effective reconstruction of latent forces demonstrated in numerical simulations.

Successful application to real-world biomedical data from the hypothalamic-pituitary-thyroid axis.

Improved estimation accuracy over existing methods.

Abstract

Latent force models, a class of hybrid modeling approaches, integrate physical knowledge of system dynamics with a latent force - an unknown, unmeasurable input modeled as a Gaussian process. In this work, we introduce two optimal state estimation frameworks to reconstruct the latent forces and to estimate the states. In contrast to state-of-the-art approaches, the designed estimators enable the consideration of system-inherent constraints. Finally, the performance of the novel frameworks is investigated in several numerical examples. In particular, we demonstrate the performance of the new framework in a real-world biomedical example - the hypothalamic-pituitary-thyroid axis - using hormone measurements.