Information value in nonparametric Dirichlet-process Gaussian-process (DPGP) mixture models

Wei, H; Lu, W; Zhu, P; Ferrari, S; Liu, M; Klein, RH; Omidshafiei, S; How, JP

Information value in nonparametric Dirichlet-process Gaussian-process (DPGP) mixture models

Wei, H Lu, W Zhu, P Ferrari, S

Liu, M Klein, RH Omidshafiei, S How, JP

Permalink

Publication Type:: Journal Article
Citation:: Automatica, 2016, 74 pp. 360 - 368
Issue Date:: 2016-12-01

Closed Access

	Filename	Description	Size
	1-s2.0-S0005109816302862-main.pdf	Published Version	1.53 MB	Adobe PDF	View/Open

Copyright Clearance Process

Recently Added
In Progress
Closed Access

This item is closed access and not available.

Full metadata record

Field	Value	Language
dc.contributor.author	Wei, H	en_US
dc.contributor.author	Lu, W	en_US
dc.contributor.author	Zhu, P	en_US
dc.contributor.author	Ferrari, S https://orcid.org/0000-0002-7652-6311	en_US
dc.contributor.author	Liu, M	en_US
dc.contributor.author	Klein, RH	en_US
dc.contributor.author	Omidshafiei, S	en_US
dc.contributor.author	How, JP https://orcid.org/0000-0001-8576-1930	en_US
dc.date.issued	2016-12-01	en_US
dc.identifier.citation	Automatica, 2016, 74 pp. 360 - 368	en_US
dc.identifier.issn	0005-1098	en_US
dc.identifier.uri	http://hdl.handle.net/10453/105635
dc.description.abstract	© 2016 Elsevier Ltd This paper presents tractable information value functions for Dirichlet-process Gaussian-process (DPGP) mixture models obtained via collocation methods and Monte Carlo integration. Quantifying information value in tractable closed form is key to solving control and estimation problems for autonomous information-gathering systems. The properties of the proposed value functions are analyzed and then demonstrated by planning sensor measurements so as to minimize the uncertainty in DPGP target models that are learned incrementally over time. Simulation results show that sensor planning based on expected KL divergence outperforms algorithms based on mutual information, particle filters, and randomized methods.	en_US
dc.relation.ispartof	Automatica	en_US
dc.relation.isbasedon	10.1016/j.automatica.2016.07.018	en_US
dc.subject.classification	Industrial Engineering & Automation	en_US
dc.title	Information value in nonparametric Dirichlet-process Gaussian-process (DPGP) mixture models	en_US
dc.type	Journal Article
utslib.citation.volume	74	en_US
utslib.for	0801 Artificial Intelligence and Image Processing	en_US
utslib.for	01 Mathematical Sciences	en_US
utslib.for	08 Information and Computing Sciences	en_US
utslib.for	09 Engineering	en_US
pubs.embargo.period	Not known	en_US
pubs.organisational-group	/University of Technology Sydney
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology
utslib.copyright.status	closed_access
pubs.publication-status	Published	en_US
pubs.volume	74	en_US

Abstract:

© 2016 Elsevier Ltd This paper presents tractable information value functions for Dirichlet-process Gaussian-process (DPGP) mixture models obtained via collocation methods and Monte Carlo integration. Quantifying information value in tractable closed form is key to solving control and estimation problems for autonomous information-gathering systems. The properties of the proposed value functions are analyzed and then demonstrated by planning sensor measurements so as to minimize the uncertainty in DPGP target models that are learned incrementally over time. Simulation results show that sensor planning based on expected KL divergence outperforms algorithms based on mutual information, particle filters, and randomized methods.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/105635