BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:The Automatic Statistician - an AI for Data Science - Zoubin Ghahr
 amani - Department of Engineering\, University of Cambridge
DTSTART:20150311T140000Z
DTEND:20150311T150000Z
UID:TALK58374@talks.cam.ac.uk
CONTACT:David Greaves
DESCRIPTION:We live an era of abundant data and there is an increasing nee
 d for methods to automate data analysis and statistics. I will describe th
 e "Automatic Statistician" (http://www.automaticstatistician.com/) \,  a p
 roject which aims to automate the exploratory analysis and modelling of da
 ta. Our approach starts by defining a large space of  related probabilisti
 c models via a grammar over models\, and then uses Bayesian marginal likel
 ihood computations 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 pe
 rformance\, and are somewhat interpretable. Our initial work has focused o
 n the learning of unknown nonparametric regression functions\, and on lear
 ning models of time series data\, both using Gaussian processes. Once a go
 od model has been found\, the Automatic Statistician generates a natural l
 anguage summary of the analysis\, producing a 10-15 page report with plots
  and tables describing the analysis.  I will discuss challenges such as: h
 ow to trade off predictive performance and interpretability\, how to trans
 late complex statistical concepts into natural language text that is under
 standable by a numerate non-statistician\, and how to integrate model chec
 king.\n\nThis is joint work with James Lloyd\, David Duvenaud\, Roger Gros
 se and Josh Tenenbaum.
LOCATION:Lecture Theatre 1\, Computer Laboratory
END:VEVENT
END:VCALENDAR