Write a Blog >>
Sat 23 Jan 2016 15:30 - 16:30 at Room St Petersburg II - Poster Session

A universal probabilistic programming language consists of a general-purpose language extended with two probabilistic features: the ability to make random (probabilistic) choices and the ability to make observations. For expressiveness and efficiency, it is useful to consider observations that have nonnegative real likelihoods rather than simple boolean truth values. A program in such a language represents a probabilistic process; the chance of producing a particular answer is determined by the random choices made along the way and the likelihoods of the observations.

Existing probabilistic programming languages typically support a constrained form of observation—such as requiring the distribution in explicit form—or a general weighting operation, like factor. This work explores the interaction between observation and the other computational features of the language. We present a big-step semantics of importance sampling with likelihood weighting for a core universal probabilistic programming language with observation propagation.

Sat 23 Jan

pps-2016
15:30 - 16:30: PPS 2016 - Poster Session at Room St Petersburg II
pps-201615:30 - 16:30
Meeting
Aleksey KligerXamarin, Inc., Sean StromstenBAE Systems, Inc.
Pre-print
pps-201615:30 - 16:30
Meeting
Theophilos GiannakopoulosBAE Systems, Inc., Mitchell WandNortheastern University, Andrew CobbNortheastern University
Pre-print
pps-201615:30 - 16:30
Meeting
Larry MossIndiana University, Chung-chieh ShanIndiana University, Alexandra SilvaRadboud University Nijmegen
Pre-print
pps-201615:30 - 16:30
Meeting
Faris Abou-SalehUniversity of Oxford, Kwok-Ho CheungUniversity of Oxford, Jeremy GibbonsUniversity of Oxford, UK
Pre-print
pps-201615:30 - 16:30
Meeting
Johannes HölzlTechnische Universität München
Pre-print
pps-201615:30 - 16:30
Meeting
Yohei MiyamotoGraduate School of Informatics, Kyoto University, Kohei Suenaga, Koji NakazawaGraduate School of Information Science, Nagoya University
Pre-print
pps-201615:30 - 16:30
Meeting
Ryan CulpepperNortheastern University
Pre-print
pps-201615:30 - 16:30
Meeting
Pre-print
pps-201615:30 - 16:30
Meeting
Adam ŚcibiorUniversity of Cambridge, Andrew D. GordonMicrosoft Research and University of Edinburgh
Pre-print
pps-201615:30 - 16:30
Meeting
Adam ŚcibiorUniversity of Cambridge, Bernhard SchölkopfMPI Tuebingen
Pre-print