Please use this identifier to cite or link to this item: http://ktisis.cut.ac.cy/handle/10488/8199
Title: A latent manifold markovian dynamics gaussian process
Authors: Chatzis, Sotirios P. 
Kosmopoulos, Dimitrios 
Keywords: Gaussian process;Stick-breaking process;Markovian dynamics;Latent manifold
Category: Mathematics
Field: Natural Sciences
Issue Date: 1-Jan-2015
Publisher: IEEE
Source: IEEE Transactions on Neural Networks and Learning Systems, 2015, Volume 26, Issue 1, Pages 70-83
Abstract: In this paper, we propose a Gaussian process model for analysis of nonlinear time series. Formulation of our model is based on the consideration that the observed data are functions of latent variables, with the associated mapping between observations and latent representations modeled through Gaussian process priors. In addition, to capture the temporal dynamics in the modeled data, we assume that subsequent latent representations depend on each other on the basis of a hidden Markov prior imposed over them. Derivation of our model is performed by marginalizing out the model parameters in closed form by using Gaussian process priors for observation mappings, and appropriate stick-breaking priors for the latent variable (Markovian) dynamics. This way, we eventually obtain a nonparametric Bayesian model for dynamical systems that accounts for uncertainty in the modeled data.We provide efficient inference algorithms for our model on the basis of a truncated variational Bayesian approximation. We demonstrate the efficacy of our approach considering a number of applications dealing with real-world data, and compare it to related state-of-the-art approaches.
URI: http://ktisis.cut.ac.cy/handle/10488/8199
ISSN: 10.1109/TNNLS.2014.2311073
2162-237X
10.1109/TNNLS.2014.2311073
DOI: 10.1109/TNNLS.2014.2311073
Rights: © IEEE
Type: Article
Appears in Collections:Άρθρα/Articles

Files in This Item:
File Description SizeFormat 
Chatzis.pdf369 kBAdobe PDFView/Open
Show full item record

SCOPUSTM   
Citations 50

2
checked on Nov 24, 2017

WEB OF SCIENCETM
Citations 20

1
Last Week
0
Last month
0
checked on Nov 22, 2017

Page view(s) 50

31
Last Week
1
Last month
2
checked on Nov 23, 2017

Download(s) 10

19
checked on Nov 23, 2017

Google ScholarTM

Check

Altmetric


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.