Fast nonparametric clustering of structured time-series

[edit]

James Hensman, University of Lancaster
Magnus Rattray, University of Manchester
Neil D. Lawrence, University of Sheffield

IEEE Transactions on Pattern Analysis and Machine Intelligence

Related Material

Abstract

In this publication, we combine two Bayesian nonparametric models: the Gaussian Process (GP) and the Dirichlet Process (DP). Our innovation in the GP model is to introduce a variation on the GP prior which enables us to model structured time-series data, i.e. data containing groups where we wish to model inter- and intra-group variability. Our innovation in the DP model is an implementation of a new fast collapsed variational inference procedure which enables us to optimize our variational approximation significantly faster than standard VB approaches. In a biological time series application we show how our model better captures salient features of the data, leading to better consistency with existing biological classifications, while the associated inference algorithm provides a significant speed-up over EM-based variational inference.


@Article{hensman-fast14,
  title = 	 {Fast nonparametric clustering of structured time-series},
  journal =  	 {IEEE Transactions on Pattern Analysis and Machine Intelligence},
  author = 	 {James Hensman and Magnus Rattray and Neil D. Lawrence},
  year = 	 {2014},
  month = 	 {00},
  edit = 	 {https://github.com/lawrennd//publications/edit/gh-pages/_posts/2014-01-01-hensman-fast14.md},
  url =  	 {http://inverseprobability.com/publications/hensman-fast14.html},
  abstract = 	 {In this publication, we combine two Bayesian nonparametric models: the Gaussian Process (GP) and the Dirichlet Process (DP). Our innovation in the GP model is to introduce a variation on the GP prior which enables us to model structured time-series data, i.e. data containing groups where we wish to model inter- and intra-group variability. Our innovation in the DP model is an implementation of a new fast collapsed variational inference procedure which enables us to optimize our variational approximation significantly faster than standard VB approaches. In a biological time series application we show how our model better captures salient features of the data, leading to better consistency with existing biological classifications, while the associated inference algorithm provides a significant speed-up over EM-based variational inference.},
  key = 	 {Hensman-fast14},
  doi = 	 {10.1109/TPAMI.2014.2318711},
  OPTgroup = 	 {}
 

}
%T Fast nonparametric clustering of structured time-series
%A James Hensman and Magnus Rattray and Neil D. Lawrence
%B 
%C IEEE Transactions on Pattern Analysis and Machine Intelligence
%D 
%F hensman-fast14
%J IEEE Transactions on Pattern Analysis and Machine Intelligence	
%P --
%R 10.1109/TPAMI.2014.2318711
%U http://inverseprobability.com/publications/hensman-fast14.html
%X In this publication, we combine two Bayesian nonparametric models: the Gaussian Process (GP) and the Dirichlet Process (DP). Our innovation in the GP model is to introduce a variation on the GP prior which enables us to model structured time-series data, i.e. data containing groups where we wish to model inter- and intra-group variability. Our innovation in the DP model is an implementation of a new fast collapsed variational inference procedure which enables us to optimize our variational approximation significantly faster than standard VB approaches. In a biological time series application we show how our model better captures salient features of the data, leading to better consistency with existing biological classifications, while the associated inference algorithm provides a significant speed-up over EM-based variational inference.
TY  - CPAPER
TI  - Fast nonparametric clustering of structured time-series
AU  - James Hensman
AU  - Magnus Rattray
AU  - Neil D. Lawrence
PY  - 2014/01/01
DA  - 2014/01/01	
ID  - hensman-fast14	
SP  - 
EP  - 
DO  - 10.1109/TPAMI.2014.2318711
UR  - http://inverseprobability.com/publications/hensman-fast14.html
AB  - In this publication, we combine two Bayesian nonparametric models: the Gaussian Process (GP) and the Dirichlet Process (DP). Our innovation in the GP model is to introduce a variation on the GP prior which enables us to model structured time-series data, i.e. data containing groups where we wish to model inter- and intra-group variability. Our innovation in the DP model is an implementation of a new fast collapsed variational inference procedure which enables us to optimize our variational approximation significantly faster than standard VB approaches. In a biological time series application we show how our model better captures salient features of the data, leading to better consistency with existing biological classifications, while the associated inference algorithm provides a significant speed-up over EM-based variational inference.
ER  -

Hensman, J., Rattray, M. & Lawrence, N.D.. (2014). Fast nonparametric clustering of structured time-series. IEEE Transactions on Pattern Analysis and Machine Intelligence :-