A Non-parametric Conditional Factor Regression Model for Multi-Dimensional Input and Response

Ava Bargi, Richard Yi Xu, Zoubin Ghahramani, Massimo Piccardi
Proceedings of the Seventeenth International Conference on Artificial Intelligence and Statistics, PMLR 33:77-85, 2014.

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.

Cite this Paper


BibTeX
@InProceedings{pmlr-v33-bargi14, title = {{A Non-parametric Conditional Factor Regression Model for Multi-Dimensional Input and Response}}, author = {Ava Bargi and Richard Yi Xu and Zoubin Ghahramani and Massimo Piccardi}, booktitle = {Proceedings of the Seventeenth International Conference on Artificial Intelligence and Statistics}, pages = {77--85}, year = {2014}, editor = {Samuel Kaski and Jukka Corander}, volume = {33}, series = {Proceedings of Machine Learning Research}, address = {Reykjavik, Iceland}, month = {22--25 Apr}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v33/bargi14.pdf}, url = {http://proceedings.mlr.press/v33/bargi14.html}, 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.} }
Endnote
%0 Conference Paper %T A Non-parametric Conditional Factor Regression Model for Multi-Dimensional Input and Response %A Ava Bargi %A Richard Yi Xu %A Zoubin Ghahramani %A Massimo Piccardi %B Proceedings of the Seventeenth International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2014 %E Samuel Kaski %E Jukka Corander %F pmlr-v33-bargi14 %I PMLR %P 77--85 %U http://proceedings.mlr.press/v33/bargi14.html %V 33 %X 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.
RIS
TY - CPAPER TI - A Non-parametric Conditional Factor Regression Model for Multi-Dimensional Input and Response AU - Ava Bargi AU - Richard Yi Xu AU - Zoubin Ghahramani AU - Massimo Piccardi BT - Proceedings of the Seventeenth International Conference on Artificial Intelligence and Statistics DA - 2014/04/02 ED - Samuel Kaski ED - Jukka Corander ID - pmlr-v33-bargi14 PB - PMLR DP - Proceedings of Machine Learning Research VL - 33 SP - 77 EP - 85 L1 - http://proceedings.mlr.press/v33/bargi14.pdf UR - http://proceedings.mlr.press/v33/bargi14.html AB - 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. ER -
APA
Bargi, A., Xu, R.Y., Ghahramani, Z. & Piccardi, M.. (2014). A Non-parametric Conditional Factor Regression Model for Multi-Dimensional Input and Response. Proceedings of the Seventeenth International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 33:77-85 Available from http://proceedings.mlr.press/v33/bargi14.html.

Related Material