Control-Data Separation and Logical Condition Propagation for Efficient Inference on Probabilistic Programs

01/05/2021
by   Ichiro Hasuo, et al.
0

We introduce a novel sampling algorithm for Bayesian inference on imperative probabilistic programs. It features a hierarchical architecture that separates control flows from data: the top-level samples a control flow, and the bottom level samples data values along the control flow picked by the top level. This separation allows us to plug various language-based analysis techniques in probabilistic program sampling; specifically, we use logical backward propagation of observations for sampling efficiency. We implemented our algorithm on top of Anglican. The experimental results demonstrate our algorithm's efficiency, especially for programs with while loops and rare observations.

READ FULL TEXT
POST COMMENT

Comments

There are no comments yet.

Authors

page 1

page 2

page 3

page 4

04/08/2021

Sound Probabilistic Inference via Guide Types

Probabilistic programming languages aim to describe and automate Bayesia...
01/08/2020

Stochastic probabilistic programs

We introduce the notion of a stochastic probabilistic program and presen...
12/22/2011

Improving the Efficiency of Approximate Inference for Probabilistic Logical Models by means of Program Specialization

We consider the task of performing probabilistic inference with probabil...
09/26/2021

Statically Bounded-Memory Delayed Sampling for Probabilistic Streams

Probabilistic programming languages aid developers performing Bayesian i...
03/01/2021

Meta-Learning an Inference Algorithm for Probabilistic Programs

We present a meta-algorithm for learning a posterior-inference algorithm...
09/10/2020

Disjunctive Delimited Control

Delimited control is a powerful mechanism for programming language exten...
06/13/2012

Sampling First Order Logical Particles

Approximate inference in dynamic systems is the problem of estimating th...
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.