BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Constructing temporal latent spaces: Representation learning for c
 lustering and imputation on time series - Vincent Fortuin\, ETH Zurich
DTSTART:20191028T110000Z
DTEND:20191028T120000Z
UID:TALK133840@talks.cam.ac.uk
CONTACT:Dr R.E. Turner
DESCRIPTION:Time series data are common in many domains\, for instance in 
 medicine or finance. Their visualization and interpretation is often cruci
 al for important decision tasks. However\, real-world time series pose man
 y challenges\, such as noise\, high dimensionality\, and missingness. Many
  time-tested machine learning models\, such as self-organizing maps (SOMs)
  and Gaussian processes (GPs)\, impose certain assumptions on their input 
 data that are not satisfied by real-world time series. We thus propose to 
 learn a representation of the time series in a latent space where these as
 sumptions are satisfied\, such that the target models can be fit to this r
 epresentation. Moreover\, one can often train the representation learner a
 nd the target model end-to-end to yield optimal performance. In this talk\
 , we are going to present two examples of this approach\, with SOMs and GP
 s as target models. We are going to lay out the conceptual background and 
 present empirical evidence on benchmark data sets and real-world medical t
 ime series.
LOCATION:Engineering Department\, CBL Room BE-438.
END:VEVENT
END:VCALENDAR