Fast Gaussian Process Based Gradient Matching for Parameter Identification in Systems of Nonlinear ODEs
Machine Learning
2019-03-04 v2 Machine Learning
Abstract
Parameter identification and comparison of dynamical systems is a challenging task in many fields. Bayesian approaches based on Gaussian process regression over time-series data have been successfully applied to infer the parameters of a dynamical system without explicitly solving it. While the benefits in computational cost are well established, a rigorous mathematical framework has been missing. We offer a novel interpretation which leads to a better understanding and improvements in state-of-the-art performance in terms of accuracy for nonlinear dynamical systems.
Cite
@article{arxiv.1804.04378,
title = {Fast Gaussian Process Based Gradient Matching for Parameter Identification in Systems of Nonlinear ODEs},
author = {Philippe Wenk and Alkis Gotovos and Stefan Bauer and Nico Gorbach and Andreas Krause and Joachim M. Buhmann},
journal= {arXiv preprint arXiv:1804.04378},
year = {2019}
}
Comments
accepted at AISTATS 2019