PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

System Identification in Gaussian Process Dynamical Systems
Ryan Turner, Marc Deisenroth and Carl Edward Rasmussen
In: Nonparametric Bayes Workshop at NIPS 2009, 12 Dec 2009, Vancouver, BC, Canada.

Abstract

The contribution of this paper is the GPIL algorithm for system identification in nonlinear dynamic systems for the special case where both the system function and the measurement function are described by Gaussian processes (GPs). Our algorithm learn GP models for both the transition function and measurement function without the necessity of ground truth observations of the latent states.

PDF - Requires Adobe Acrobat Reader or other PDF viewer.
EPrint Type:Conference or Workshop Item (Poster)
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Learning/Statistics & Optimisation
Theory & Algorithms
ID Code:6483
Deposited By:Marc Deisenroth
Deposited On:08 March 2010