BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Typed functional probabilistic programming: ready for practical us
 e? - Tom Nielsen\, OpenBrain
DTSTART:20131211T100000Z
DTEND:20131211T104500Z
UID:TALK49304@talks.cam.ac.uk
CONTACT:Microsoft Research Cambridge Talks Admins
DESCRIPTION:Advances in inference algorithms and probability theory have s
 purred recent interest in probabilistic programming. However\, it is still
  not clear which design leads to the maximal flexibility in terms of explo
 iting probabilistic inference. In this talk\, I will motivate the need for
  and potential of probabilistic programming and describe some of the\nlimi
 tations of current approaches. The central claim of the talk is that combi
 ning functional and probabilistic programming\, and in particular exploiti
 ng the fact that probability distributions are monads\, leads to a very ex
 pressive language for probabilistic models. I will define the handful of o
 perators that need to be added to a purely functional programming language
  in order to achieve practically useful Bayesian inference and allow flexi
 ble post-inference computations. Adding a\ndifferential operator extends t
 he paradigm to inference in models based on ordinary and stochastic differ
 ential equations. Static types can then introduce new guarantees about the
  propagation and maintenance of uncertainty. Finally\, I will show how thi
 s core computational platform can be integrated with more traditional comp
 uter applications.
LOCATION:Auditorium\, Microsoft Research Ltd\, 21 Station Road\, Cambridge
 \, CB1 2FB
END:VEVENT
END:VCALENDAR