BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:A compositional approach to scalable statistical modelling and com
 putation - Darren Wilkinson (Newcastle University)
DTSTART:20180208T110000Z
DTEND:20180208T120000Z
UID:TALK100723@talks.cam.ac.uk
CONTACT:INI IT
DESCRIPTION:In statistics\, and in life\, we typically solve big problems 
 by (recursively) breaking them down into smaller problems that we can solv
 e more easily\, and then compose the solutions of the smaller problems to 
 provide a solution to the big problem that we are really interested in. Th
 is "divide and conquer" approach is necessary for the development of genui
 nely scalable models and algorithms. It is therefore unfortunate that stat
 istical models and algorithms are not usually formulated in a composable w
 ay\, and that the programming languages typically used for scientific and 
 statistical computing also fail to naturally support composition of models
 \, data and computation. The mathematical subject of category theory is in
  many ways the mathematical study of composition\, and provides significan
 t insight into the development of more compositional models of computation
 . Functional programming languages which are strongly influenced by catego
 ry theory turn out to be much better suited to the development of scalable
  statistical algorithms than the imperative programming languages more com
 monly used. Expressing algorithms in a functional/categorical way is not o
 nly more elegant\, concise and less error-prone\, but provides numerous mo
 re tangible benefits\, such as automatic parallelisation and distribution 
 of algorithms. I will illustrate the concepts using examples such as the s
 tatistical analysis of streaming data\, image analysis\, numerical integra
 tion of PDEs\, particle filtering\, Gibbs sampling\, and probabilistic pro
 gramming\, using concepts from category theory such as functors\, monads a
 nd comonads. Illustrative code snippets will given using the Scala program
 ming language.
LOCATION:Seminar Room 2\, Newton Institute
END:VEVENT
END:VCALENDAR