Probabilistic programming Conference Paper

Author(s): Gordon, Andrew D; Henzinger, Thomas A; Nori, Aditya V; Rajamani, Sriram K
Title: Probabilistic programming
Affiliation IST Austria
Abstract: Probabilistic programs are usual functional or imperative programs with two added constructs: (1) the ability to draw values at random from distributions, and (2) the ability to condition values of variables in a program via observations. Models from diverse application areas such as computer vision, coding theory, cryptographic protocols, biology and reliability analysis can be written as probabilistic programs. Probabilistic inference is the problem of computing an explicit representation of the probability distribution implicitly specified by a probabilistic program. Depending on the application, the desired output from inference may vary-we may want to estimate the expected value of some function f with respect to the distribution, or the mode of the distribution, or simply a set of samples drawn from the distribution. In this paper, we describe connections this research area called \Probabilistic Programming" has with programming languages and software engineering, and this includes language design, and the static and dynamic analysis of programs. We survey current state of the art and speculate on promising directions for future research.
Keywords: program analysis; Machine learning; Probabilistic programming
Conference Title: FOSE: Future of Software Engineering
Conference Dates: May 31 - June 7, 2014
ISBN: 978-145032865-4
Publisher: ACM  
Date Published: 2014-05-31
Start Page: 167
End Page: 181
Sponsor: This work was supported in part by the ERC Advanced Grant QUAREM (Quantitative Reactive Modeling) and the FWF NFN RiSE (Rigorous Systems Engineering).
DOI: 10.1145/2593882.2593900
Notes: We are grateful to Johannes Borgström, Audris Mockus, Dan Suciu, and Marcin Szymczak for feedback on a draft.
Open access: no
IST Austria Authors
  1. Thomas A. Henzinger
    415 Henzinger
Related IST Austria Work