Probabilistic programming (PP) is a programming paradigm in which probabilistic models are specified and inference for these models is performed automatically.[1] It represents an attempt to unify probabilistic modeling and traditional general purpose programming in order to make the former easier and more widely applicable.[2][3] It can be used to create systems that help make decisions in the face of uncertainty.
Programming languages used for probabilistic programming are referred to as "probabilistic programming languages" (PPLs).
Applicationsedit
Probabilistic reasoning has been used for a wide variety of tasks such as predicting stock prices, recommending movies, diagnosing computers, detecting cyber intrusions and image detection.[4] However, until recently (partially due to limited computing power), probabilistic programming was limited in scope, and most inference algorithms had to be written manually for each task.
Nevertheless, in 2015, a 50-line probabilistic computer vision program was used to generate 3D models of human faces based on 2D images of those faces. The program used inverse graphics as the basis of its inference method, and was built using the Picture package in Julia.[4] This made possible "in 50 lines of code what used to take thousands".[5][6]
The Gen probabilistic programming library (also written in Julia) has been applied to vision and robotics tasks.[7]
More recently, the probabilistic programming system Turing.jl has been applied in various pharmaceutical[8] and economics applications.[9]
Probabilistic programming in Julia has also been combined with differentiable programming by combining the Julia package Zygote.jl with Turing.jl. [10]
Probabilistic programming languages are also commonly used in Bayesian cognitive science to develop and evaluate models of cognition. [11]
Probabilistic programming languagesedit
PPLs often extend from a basic language. The choice of underlying basic language depends on the similarity of the model to the basic language's ontology, as well as commercial considerations and personal preference. For instance, Dimple[12] and Chimple[13] are based on Java, Infer.NET is based on .NET Framework,[14] while PRISM extends from Prolog.[15] However, some PPLs such as WinBUGS offer a self-contained language, that maps closely to the mathematical representation of the statistical models, with no obvious origin in another programming language.[16][17]
The language for winBUGS was implemented to perform Bayesian computation using Gibbs Sampling (and related algorithms). Although implemented in a relatively unknown programming language (Component Pascal), this language permits Bayesian inference for a wide variety of statistical models using a flexible computational approach. The same BUGS language may be used to specify Bayesian models for inference via different computational choices ("samplers") and conventions or defaults, using a standalone program winBUGS (or related R packages, rbugs and r2winbugs) and JAGS (Just Another Gibbs Sampler, another standalone program with related R packages including rjags, R2jags, and runjags). More recently, other languages to support Bayesian model specification and inference allow different or more efficient choices for the underlying Bayesian computation, and are accessible from the R data analysis and programming environment, e.g.: Stan, NIMBLE and NUTS. The influence of the BUGS language is evident in these later languages, which even use the same syntax for some aspects of model specification.
Several PPLs are in active development, including some in beta test. Two popular tools are Stan and PyMC.[18]
Relationaledit
A probabilistic relational programming language (PRPL) is a PPL specially designed to describe and infer with probabilistic relational models (PRMs).
A PRM is usually developed with a set of algorithms for reducing, inference about and discovery of concerned distributions, which are embedded into the corresponding PRPL.
Most approaches to probabilistic logic programming are based on the distribution semantics, which splits a program into a set of probabilistic facts and a logic program. It defines a probability distribution on interpretations of the Herbrand universe of the program.[19]
List of probabilistic programming languagesedit
This list summarises the variety of PPLs that are currently available, and clarifies their origins.
Reasoning about variables as probability distributions causes difficulties for novice programmers, but these difficulties can be addressed through use of Bayesian network visualisations and graphs of variable distributions embedded within the source code editor.[71]
^ abc"Short probabilistic programming machine-learning code replaces complex programs for computer-vision tasks". KurzweilAI. April 13, 2015. Retrieved November 27, 2017.
^Hardesty, Larry (April 13, 2015). "Graphics in reverse".
^"MIT shows off machine-learning script to make CREEPY HEADS". The Register.
^"MIT's Gen programming system flattens the learning curve for AI projects". VentureBeat. June 27, 2019. Retrieved June 27, 2019.
^Semenova, Elizaveta; Williams, Dominic P.; Afzal, Avid M.; Lazic, Stanley E. (November 1, 2020). "A Bayesian neural network for toxicity prediction". Computational Toxicology. 16: 100133. doi:10.1016/j.comtox.2020.100133. ISSN 2468-1113. S2CID 225362130.
^Williams, Dominic P.; Lazic, Stanley E.; Foster, Alison J.; Semenova, Elizaveta; Morgan, Paul (2020), "Predicting Drug-Induced Liver Injury with Bayesian Machine Learning", Chemical Research in Toxicology, 33 (1): 239–248, doi:10.1021/acs.chemrestox.9b00264, PMID 31535850, S2CID 202689667
^Innes, Mike; Edelman, Alan; Fischer, Keno; Rackauckas, Chris; Saba, Elliot; Viral B Shah; Tebbutt, Will (2019). "∂P: A Differentiable Programming System to Bridge Machine Learning and Scientific Computing". arXiv:1907.07587 [cs.PL].
^Goodman, Noah D; Tenenbaum, Joshua B; Buchsbaum, Daphna; Hartshorne, Joshua; Hawkins, Robert; O'Donnell, Timothy J; Tessler, Michael Henry. "Probabilistic Models of Cognition". Probabilistic Models of Cognition - 2nd Edition. Retrieved May 27, 2023.
^"Bean Machine - A universal probabilistic programming language to enable fast and accurate Bayesian analysis". beanmachine.org.
^"Probabilistic Programming with CuPPL". popl19.sigplan.org.
^Collins, Alexander; Grewe, Dominik; Grover, Vinod; Lee, Sean; Susnea, Adriana (June 9, 2014). "NOVA: A Functional Language for Data Parallelism". Proceedings of ACM SIGPLAN International Workshop on Libraries, Languages, and Compilers for Array Programming. Array'14. pp. 8–13. doi:10.1145/2627373.2627375. ISBN 9781450329378. S2CID 6748967. {{cite book}}: |work= ignored (help)
^"Venture -- a general-purpose probabilistic programming platform". mit.edu. Archived from the original on January 25, 2016. Retrieved September 20, 2014.
^"Probabilistic C". ox.ac.uk. Archived from the original on January 4, 2016. Retrieved March 24, 2015.
^"The Anglican Probabilistic Programming System". ox.ac.uk. January 6, 2021.
^"IBAL Home Page". Archived from the original on December 26, 2010.
^"BayesDB on SQLite. A Bayesian database table for querying the probable implications of data as easily as SQL databases query the data itself". GitHub. December 26, 2021.
^"Bayesian Logic (BLOG)". mit.edu. Archived from the original on June 16, 2011.
^"diff-SAT (probabilistic SAT/ASP)". GitHub. October 8, 2021.
^Dey, Debabrata; Sarkar, Sumit (1998). "PSQL: A query language for probabilistic relational data". Data & Knowledge Engineering. 28: 107–120. doi:10.1016/S0169-023X(98)00015-9.
^"Factorie - Probabilistic programming with imperatively-defined factor graphs - Google Project Hosting". google.com.
^"PMTK3 - probabilistic modeling toolkit for Matlab/Octave, version 3 - Google Project Hosting". google.com.
^ProbaYes. "ProbaYes - Ensemble, nous valorisations vos données". probayes.com. Archived from the original on March 5, 2016. Retrieved November 26, 2013.
^Let's Chance: Playful Probabilistic Programming for Children | Extended Abstracts of the 2020 CHI Conference on Human Factors in Computing Systems. Chi Ea '20. April 25, 2020. pp. 1–7. doi:10.1145/3334480.3383071. ISBN 9781450368193. S2CID 216079395. Retrieved August 1, 2020. {{cite book}}: |website= ignored (help)
^"The Turing language for probabilistic programming". GitHub. December 28, 2021.
^"Gen: A General Purpose Probabilistic Programming Language with Programmable Inference". Retrieved June 17, 2019.
^"LF-PPL: A Low-Level First Order Probabilistic Programming Language for Non-Differentiable Models". ox.ac.uk. November 2, 2019.
^"Troll dice roller and probability calculator". topps.diku.dk.
^"Edward – Home". edwardlib.org. Retrieved January 17, 2017.
^Perov, Yura; Graham, Logan; Gourgoulias, Kostis; Richens, Jonathan G.; Lee, Ciarán M.; Baker, Adam; Johri, Saurabh (January 28, 2020), MultiVerse: Causal Reasoning using Importance Sampling in Probabilistic Programming, arXiv:1910.08091
^Gorinova, Maria I.; Sarkar, Advait; Blackwell, Alan F.; Syme, Don (January 1, 2016). "A Live, Multiple-Representation Probabilistic Programming Environment for Novices". Proceedings of the 2016 CHI Conference on Human Factors in Computing Systems. CHI '16. New York, NY, USA: ACM. pp. 2533–2537. doi:10.1145/2858036.2858221. ISBN 9781450333627. S2CID 3201542.
External linksedit
List of Probabilistic Model Mini Language Toolkits