BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Probabilistic Dimensional Reduction with the Gaussian Process Late
 nt Variable Model - Dr Neil Lawrence\, Computer Science\, University of Sh
 effield
DTSTART:20060307T123000Z
DTEND:20060307T133000Z
UID:TALK4822@talks.cam.ac.uk
CONTACT:Zoubin Ghahramani
DESCRIPTION:                 Density modelling in high dimensions is a\n  
                very difficult problem. Traditional approaches\, such\n    
              as mixtures of Gaussians\, typically fail to capture\n       
           the structure of data sets in high dimensional\n                
  spaces. In this talk we will argue that for many data\n                 s
 ets of interest\, the data can be represented as a\n                 lower
  dimensional manifold immersed in the higher\n                 dimensional
  space. We will then present the Gaussian\n                 Process Latent
  Variable Model (GP-LVM)\, a non-linear\n                 probabilistic va
 riant of principal component analysis\n                 (PCA) which implic
 itly assumes that the data lies on\n                 a lower dimensional s
 pace.\n\n                 Having introduced the GP-LVM we will review\n   
               extensions to the algorithm\, including dynamics\,\n        
          learning of large data sets and back constraints.  We\n          
        will demonstrate the application of the model and its\n            
      extensions to a range of data sets\, including human\n               
   motion data\, a vowel data set and a robot mapping\n                 pro
 blem.\n
LOCATION:LR10\, Engineering\, Department of
END:VEVENT
END:VCALENDAR