Algorithmic Linearly Constrained Gaussian Processes

Abstract

We algorithmically construct multi-output Gaussian process priors which satisfy linear differential equations. Our approach attempts to parametrize all solutions of the equations using Gröbner bases. If successful, a push forward Gaussian process along the paramerization is the desired prior. We consider several examples from physics, geomathematics and control, among them the full inhomogeneous system of Maxwell's equations. By bringing together stochastic learning and computer algebra in a novel way, we combine noisy observations with precise algebraic computations.

NIPS 2018

References in corpus (3)

• Deep Neural Networks as Gaussian Processes
• Involutive Algorithms for Computing Groebner Bases
• Linearly constrained Gaussian processes

Cited by in corpus (5)

• Linearly Constrained Neural Networks
• Designing Robust Biotechnological Processes Regarding Variabilities using Multi-Objective Optimization Applied to a Biopharmaceutical Seed Train Design
• Linearly Constrained Gaussian Processes with Boundary Conditions
• Covariance models and Gaussian process regression for the wave equation. Application to related inverse problems
• Stochastic Processes Under Linear Differential Constraints : Application to Gaussian Process Regression for the 3 Dimensional Free Space Wave Equation