基于控制数据分离和逻辑条件传播的概率程序高效推理

IF 0.7 4区数学 Q3 COMPUTER SCIENCE, THEORY & METHODS

Journal of Logical and Algebraic Methods in Programming Pub Date : 2023-10-05 DOI:10.1016/j.jlamp.2023.100922

Ichiro Hasuo , Yuichiro Oyabu , Clovis Eberhart , Kohei Suenaga , Kenta Cho , Shin-ya Katsumata

{"title":"基于控制数据分离和逻辑条件传播的概率程序高效推理","authors":"Ichiro Hasuo , Yuichiro Oyabu , Clovis Eberhart , Kohei Suenaga , Kenta Cho , Shin-ya Katsumata","doi":"10.1016/j.jlamp.2023.100922","DOIUrl":null,"url":null,"abstract":"<div><p>We present a novel sampling framework for probabilistic programs. The framework combines two recent ideas—<em>control-data separation</em> and <em>logical condition propagation</em>—in a nontrivial manner so that the two ideas boost the benefits of each other. 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.</p></div>","PeriodicalId":48797,"journal":{"name":"Journal of Logical and Algebraic Methods in Programming","volume":"136 ","pages":"Article 100922"},"PeriodicalIF":0.7000,"publicationDate":"2023-10-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Control-data separation and logical condition propagation for efficient inference on probabilistic programs\",\"authors\":\"Ichiro Hasuo , Yuichiro Oyabu , Clovis Eberhart , Kohei Suenaga , Kenta Cho , Shin-ya Katsumata\",\"doi\":\"10.1016/j.jlamp.2023.100922\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><p>We present a novel sampling framework for probabilistic programs. The framework combines two recent ideas—<em>control-data separation</em> and <em>logical condition propagation</em>—in a nontrivial manner so that the two ideas boost the benefits of each other. 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.</p></div>\",\"PeriodicalId\":48797,\"journal\":{\"name\":\"Journal of Logical and Algebraic Methods in Programming\",\"volume\":\"136 \",\"pages\":\"Article 100922\"},\"PeriodicalIF\":0.7000,\"publicationDate\":\"2023-10-05\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Journal of Logical and Algebraic Methods in Programming\",\"FirstCategoryId\":\"94\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S2352220823000767\",\"RegionNum\":4,\"RegionCategory\":\"数学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q3\",\"JCRName\":\"COMPUTER SCIENCE, THEORY & METHODS\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Logical and Algebraic Methods in Programming","FirstCategoryId":"94","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2352220823000767","RegionNum":4,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"COMPUTER SCIENCE, THEORY & METHODS","Score":null,"Total":0}

引用次数: 0

摘要

我们提出了一种新的概率规划抽样框架。该框架以一种非凡的方式结合了两种最新的思想——控制-数据分离和逻辑条件传播，从而使这两种思想相互促进。我们在英国国教的基础上实现了我们的算法。实验结果证明了该算法的有效性，特别是对于具有while循环和罕见观测值的程序。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

Control-data separation and logical condition propagation for efficient inference on probabilistic programs

We present a novel sampling framework for probabilistic programs. The framework combines two recent ideas—control-data separation and logical condition propagation—in a nontrivial manner so that the two ideas boost the benefits of each other. 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.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

Journal of Logical and Algebraic Methods in Programming COMPUTER SCIENCE, THEORY & METHODS-LOGIC

CiteScore

2.60

自引率

22.20%

发文量

期刊介绍： The Journal of Logical and Algebraic Methods in Programming is an international journal whose aim is to publish high quality, original research papers, survey and review articles, tutorial expositions, and historical studies in the areas of logical and algebraic methods and techniques for guaranteeing correctness and performability of programs and in general of computing systems. All aspects will be covered, especially theory and foundations, implementation issues, and applications involving novel ideas.