BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:An Automated Statistician which learns Bayesian nonparametric mode
 ls of time series data - Ghahramani\, Z (University of Cambridge)
DTSTART:20140116T141500Z
DTEND:20140116T150000Z
UID:TALK49977@talks.cam.ac.uk
CONTACT:Mustapha Amrani
DESCRIPTION:I will describe the "Automated Statistician"\, a project which
  aims to automate the exploratory analysis and modelling of data. Our appr
 oach starts by defining a large space of related probabilistic models via 
 a grammar over models\, and then uses Bayesian marginal likelihood computa
 tions to search over this space for one or a few good models of the data. 
 The aim is to find models which have both good predictive performance\, an
 d are somewhat interpretable. Our initial work has focused on the learning
  of unknown nonparametric regression functions\, and on learning models of
  time series data\, both using Gaussian processes. Once a good model has b
 een found\, the Automated Statistician generates a natural language summar
 y of the analysis\, producing a 10-15 page report with plots and tables de
 scribing the analysis. I will focus in particular on the modelling of time
  series\, including how we handle change points in Gaussian process models
 . I will also discuss challenges su ch as: how to trade off predictive per
 formance and interpretability\, how to translate complex statistical conce
 pts into natural language text that is understandable by a numerate non-st
 atistician\, and how to integrate model checking.\n \nThis is joint work w
 ith James Lloyd and David Duvenaud (Cambridge) and Roger Grosse and Josh T
 enenbaum (MIT).\n 
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR