Linear Latent Force Models Using Gaussian Processes

Abstract

Purely data driven approaches for machine learning present difficulties when data is scarce relative to the complexity of the model or when the model is forced to extrapolate. On the other hand, purely mechanistic approaches need to identify and specify all the interactions in the problem at hand (which may not be feasible) and still leave the issue of how to parameterize the system. In this paper, we present a hybrid approach using Gaussian processes and differential equations to combine data driven modelling with a physical model of the system. We show how different, physically-inspired, kernel functions can be developed through sensible, simple, mechanistic assumptions about the underlying system. The versatility of our approach is illustrated with three case studies from motion capture, computational biology and geostatistics.

Definition and Composition of Motor Primitives Using Latent Force Models and Hidden Markov Models

“…In order to keep the model flexible enough to fit arbitrary trajectories even under circumstances where the mechanistic assumptions are not rigorous fulfilled [3], a forcing term is added and it is shown in the right side of equation (1). This forcing term is governed by a set of Q latent functions {u q (t)} Q q=1 whose contribution to the outputs dynamics is regulated by a set of constants {S d,q } which are known as the sensitivities.…”

Risk Assessment of Power System Security based on a Hybrid Optimization GP Method

“…Gaussian Process (GP) [14], is also known as the normal random process, defined as a collection of random variables. Any finite number of random variables in the collection obwy joint Gaussian distribution, that is to say, for any set of random variables X and corresponding to the process state () fX, the joint probability distribution of them follows n -dimensional Gaussian distribution.…”

Latent Variable Analysis of Causal Interactions in Atrial Fibrillation Electrograms

“…In this paper, we use an advanced machine learning model, the Gaussian process latent force model (GP-LFM) [9][10][11], to describe the intracardiac electrocardiograms (a.k.a. electrograms (EGMs)) acquired during RF ablation at the electrophisiology laboratory.…”

“…However, no consensus has been attained yet on which areas should be ablated, the success rate of a single procedure is still unsatisfactory, and its relative effectiveness w.r.t. the use of antiarrhythmic drugs remains controversial [5][6][7][8].In this paper, we use an advanced machine learning model, the Gaussian process latent force model (GP-LFM) [9][10][11], to describe the intracardiac electrocardiograms (a.k.a. electrograms (EGMs)) acquired during RF ablation at the electrophisiology laboratory.…”