The Beta-Bernoulli process and algebraic effects
Programming Languages
2018-09-11 v2
Abstract
In this paper we use the framework of algebraic effects from programming language theory to analyze the Beta-Bernoulli process, a standard building block in Bayesian models. Our analysis reveals the importance of abstract data types, and two types of program equations, called commutativity and discardability. We develop an equational theory of terms that use the Beta-Bernoulli process, and show that the theory is complete with respect to the measure-theoretic semantics, and also in the syntactic sense of Post. Our analysis has a potential for being generalized to other stochastic processes relevant to Bayesian modelling, yielding new understanding of these processes from the perspective of programming.
Cite
@article{arxiv.1802.09598,
title = {The Beta-Bernoulli process and algebraic effects},
author = {Sam Staton and Dario Stein and Hongseok Yang and Nathanael L. Ackerman and Cameron E. Freer and Daniel M. Roy},
journal= {arXiv preprint arXiv:1802.09598},
year = {2018}
}
Comments
To appear in Proc. ICALP 2018