A non-parametric conditional factor regression model for multi-dimensional input and response

Bargi, A; Da Xu, RY; Ghahramani, Z; Piccardi, M

A non-parametric conditional factor regression model for multi-dimensional input and response

Bargi, A Da Xu, RY Ghahramani, Z Piccardi, M

Permalink

Publication Type:: Conference Proceeding
Citation:: Journal of Machine Learning Research, 2014, 33 pp. 77 - 85
Issue Date:: 2014-01-01

Open Access

Copyright Clearance Process

Recently Added
In Progress
Open Access

This item is open access.

Adobe PDF

Download Accepted Manuscript versionAdobe PDF (497.24 kB)

View statistics

Full metadata record

Field	Value	Language
dc.contributor.author	Bargi, A	en_US
dc.contributor.author	Da Xu, RY	en_US
dc.contributor.author	Ghahramani, Z	en_US
dc.contributor.author	Piccardi, M https://orcid.org/0000-0001-9250-6604	en_US
dc.date.issued	2014-01-01	en_US
dc.identifier.citation	Journal of Machine Learning Research, 2014, 33 pp. 77 - 85	en_US
dc.identifier.issn	1532-4435	en_US
dc.identifier.uri	http://hdl.handle.net/10453/35004
dc.identifier.uri	http://hdl.handle.net/10453/35123
dc.identifier.uri	http://hdl.handle.net/10453/35122
dc.identifier.uri	http://hdl.handle.net/10453/35121
dc.identifier.uri	http://hdl.handle.net/10453/35120
dc.description.abstract	In this paper, we propose a non-parametric conditional factor regression (NCFR) model for domains with multi-dimensional input and response. NCFR enhances linear regression in two ways: a) introducing low-dimensional latent factors leading to dimensionality reduction and b) integrating the Indian Buffet Process as prior for the latent layer to dynamically derive an optimal number of sparse factors. Thanks to IBP's enhancements to the latent factors, NCFR can significantly avoid over-fitting even in the case of a very small sample size compared to the dimensionality. Experimental results on three diverse datasets comparing NCRF to a few baseline alternatives give evidence of its robust learning, remarkable predictive performance, good mixing and computational efficiency.	en_US
dc.relation.ispartof	Journal of Machine Learning Research	en_US
dc.subject.classification	Artificial Intelligence & Image Processing	en_US
dc.title	A non-parametric conditional factor regression model for multi-dimensional input and response	en_US
dc.type	Conference Proceeding
utslib.citation.volume	33	en_US
utslib.for	080104 Computer Vision	en_US
utslib.for	080109 Pattern Recognition and Data Mining	en_US
utslib.for	08 Information and Computing Sciences	en_US
utslib.for	17 Psychology and Cognitive Sciences	en_US
dc.location.activity	Reykjavik, Iceland
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
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology/School of Electrical and Data Engineering
pubs.organisational-group	/University of Technology Sydney/Strength - GBDTC - Global Big Data Technologies
pubs.organisational-group	/University of Technology Sydney/Strength - INEXT - Innovation in IT Services and Applications
utslib.copyright.status	open_access
pubs.publication-status	Published	en_US
pubs.volume	33	en_US

Abstract:

In this paper, we propose a non-parametric conditional factor regression (NCFR) model for domains with multi-dimensional input and response. NCFR enhances linear regression in two ways: a) introducing low-dimensional latent factors leading to dimensionality reduction and b) integrating the Indian Buffet Process as prior for the latent layer to dynamically derive an optimal number of sparse factors. Thanks to IBP's enhancements to the latent factors, NCFR can significantly avoid over-fitting even in the case of a very small sample size compared to the dimensionality. Experimental results on three diverse datasets comparing NCRF to a few baseline alternatives give evidence of its robust learning, remarkable predictive performance, good mixing and computational efficiency.

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

http://hdl.handle.net/10453/35004