Home

About

Advanced Search

Browse by Discipline

Scientific Societies

E-print Alerts

Add E-prints

E-print Network
FAQHELPSITE MAPCONTACT US


  Advanced Search  

 
Archipelago: Nonparametric Bayesian Semi-Supervised Learning Ryan Prescott Adams rpa23@cam.ac.uk
 

Summary: Archipelago: Nonparametric Bayesian Semi-Supervised Learning
Ryan Prescott Adams rpa23@cam.ac.uk
Cavendish Laboratory, University of Cambridge, Cambridge CB3 0HE, UK
Zoubin Ghahramani zoubin@eng.cam.ac.uk
Engineering Department, University of Cambridge, Cambridge CB2 1PZ, UK
Machine Learning Department, Carnegie Mellon University, Pittsburgh, PA 15213
Abstract
Semi-supervised learning (SSL), is classifica-
tion where additional unlabeled data can be
used to improve accuracy. Generative ap-
proaches are appealing in this situation, as
a model of the data's probability density can
assist in identifying clusters. Nonparametric
Bayesian methods, while ideal in theory due
to their principled motivations, have been dif-
ficult to apply to SSL in practice. We present
a nonparametric Bayesian method that uses
Gaussian processes for the generative model,
avoiding many of the problems associated
with Dirichlet process mixture models. Our

  

Source: Adams, Ryan Prescott - Department of Electrical and Computer Engineering, University of Toronto
Edinburgh, University of - Division of Informatics, Institute for Adaptive and Neural Computation

 

Collections: Computer Technologies and Information Sciences