λPSI: exact inference for higher-order probabilistic programs
Authors:
Open Access
- 11 June 2020
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 883-897
Abstract
We present λPSI, the first probabilistic programming language and system that supports higher-order exact inference for probabilistic programs with first-class functions, nested inference and discrete, continuous and mixed random variables. λPSI’s solver is based on symbolic reasoning and computes the exact distribution represented by a program. We show that λPSI is practically effective—it automatically computes exact distributions for a number of interesting applications, from rational agents to information theory, many of which could so far only be handled approximately.Keywords
Funding Information
- European Research Council (Starting Grant 680358)
This publication has 11 references indexed in Scilit:
- Formal verification of higher-order probabilistic programs: reasoning about approximation, convergence, Bayesian inference, and optimizationProceedings of the ACM on Programming Languages, 2019
- Denotational validation of higher-order Bayesian inferenceProceedings of the ACM on Programming Languages, 2017
- Symbolic conditioning of arrays in probabilistic programsProceedings of the ACM on Programming Languages, 2017
- Commutative Semantics for Probabilistic Programming2017
- Coupling proofs are probabilistic product programs2017
- PSI: Exact Symbolic Inference for Probabilistic Programs2016
- Semantics for probabilistic programming2016
- Probabilistic Inference by Program Transformation in Hakaru (System Description)2016
- The information content in sorting algorithms2012
- Estimation of Entropy and Mutual InformationNeural Computation, 2003