Control-data separation and logical condition propagation for efficient inference on probabilistic programs
From MaRDI portal
Publication:6151609
DOI10.1016/J.JLAMP.2023.100922arXiv2101.01502OpenAlexW4387374098MaRDI QIDQ6151609FDOQ6151609
Authors: Ichiro Hasuo, Yuichiro Oyabu, Clovis Eberhart, Kohei Suenaga, Kenta Cho, Shin-ya Katsumata 
Publication date: 12 February 2024
Published in: Journal of Logical and Algebraic Methods in Programming (Search for Journal in Brave)
Abstract: 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.
Full work available at URL: https://arxiv.org/abs/2101.01502
Cites Work
- Title not available (Why is that?)
- Title not available (Why is that?)
- The Continuum-Armed Bandit Problem
- Regret Analysis of Stochastic and Nonstochastic Multi-armed Bandit Problems
- Title not available (Why is that?)
- Semantics of probabilistic programs
- Introduction to Multi-Armed Bandits
- Semantics for probabilistic programming
Cited In (1)
This page was built for publication: Control-data separation and logical condition propagation for efficient inference on probabilistic programs
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6151609)
