Hybrid Discriminative-Generative Approaches with Gaussian Processes

[edit]

Ricardo Andrade-Pacheco, UCSF Global Health Science
James Hensman, University of Lancaster
Neil D. Lawrence, University of Sheffield

in Proceedings of the Seventeenth International Workshop on Artificial Intelligence and Statistics 33, pp 47-56

Related Material

Abstract

Machine learning practitioners are often faced with a choice between a discriminative and a generative approach to modelling. Here, we present a model based on a hybrid approach that breaks down some of the barriers between the discriminative and generative points of view, allowing continuous dimensionality reduction of hybrid discrete-continous data, discriminative classification with missing inputs and manifold learning informed by class labels.


@InProceedings{andrade-hybrid14,
  title = 	 {Hybrid Discriminative-Generative Approaches with Gaussian Processes},
  author = 	 {Ricardo Andrade-Pacheco and James Hensman and Neil D. Lawrence},
  booktitle = 	 {Proceedings of the Seventeenth International Workshop on Artificial Intelligence and Statistics},
  pages = 	 {47},
  year = 	 {2014},
  editor = 	 {Sami Kaski and Jukka Corander},
  volume = 	 {33},
  address = 	 {Iceland},
  month = 	 {00},
  publisher = 	 {JMLR W\&CP 33},
  edit = 	 {https://github.com/lawrennd//publications/edit/gh-pages/_posts/2014-01-01-andrade-hybrid14.md},
  url =  	 {http://inverseprobability.com/publications/andrade-hybrid14.html},
  abstract = 	 {Machine learning practitioners are often faced with a choice between a discriminative and a generative approach to modelling. Here, we present a model based on a hybrid approach that breaks down some of the barriers between the discriminative and generative points of view, allowing continuous dimensionality reduction of hybrid discrete-continous data, discriminative classification with missing inputs and manifold learning informed by class labels.},
  crossref =  {Kaski:aistats14},
  key = 	 {Andrade:hybrid14},
  linkpdf = 	 {http://jmlr.org/proceedings/papers/v33/andradepacheco14.pdf},
  linksoftware = {https://github.com/SheffieldML/GPy},
  OPTgroup = 	 {}
 

}
%T Hybrid Discriminative-Generative Approaches with Gaussian Processes
%A Ricardo Andrade-Pacheco and James Hensman and Neil D. Lawrence
%B 
%C Proceedings of the Seventeenth International Workshop on Artificial Intelligence and Statistics
%D 
%E Sami Kaski and Jukka Corander
%F andrade-hybrid14
%I JMLR W\&CP 33	
%P 47--56
%R 
%U http://inverseprobability.com/publications/andrade-hybrid14.html
%V 33
%X Machine learning practitioners are often faced with a choice between a discriminative and a generative approach to modelling. Here, we present a model based on a hybrid approach that breaks down some of the barriers between the discriminative and generative points of view, allowing continuous dimensionality reduction of hybrid discrete-continous data, discriminative classification with missing inputs and manifold learning informed by class labels.
TY  - CPAPER
TI  - Hybrid Discriminative-Generative Approaches with Gaussian Processes
AU  - Ricardo Andrade-Pacheco
AU  - James Hensman
AU  - Neil D. Lawrence
BT  - Proceedings of the Seventeenth International Workshop on Artificial Intelligence and Statistics
PY  - 2014/01/01
DA  - 2014/01/01
ED  - Sami Kaski
ED  - Jukka Corander	
ID  - andrade-hybrid14
PB  - JMLR W\&CP 33	
SP  - 47
EP  - 56
L1  - http://jmlr.org/proceedings/papers/v33/andradepacheco14.pdf
UR  - http://inverseprobability.com/publications/andrade-hybrid14.html
AB  - Machine learning practitioners are often faced with a choice between a discriminative and a generative approach to modelling. Here, we present a model based on a hybrid approach that breaks down some of the barriers between the discriminative and generative points of view, allowing continuous dimensionality reduction of hybrid discrete-continous data, discriminative classification with missing inputs and manifold learning informed by class labels.
ER  -

Andrade-Pacheco, R., Hensman, J. & Lawrence, N.D.. (2014). Hybrid Discriminative-Generative Approaches with Gaussian Processes. Proceedings of the Seventeenth International Workshop on Artificial Intelligence and Statistics 33:47-56