POSTER: Fast Parallel Exact Inference on Bayesian Networks

Jiantong Jiang; Zeyi Wen; Atif Mansoor; Ajmal Mian

doi:10.1145/3572848.3577476

POSTER: Fast Parallel Exact Inference on Bayesian Networks

Jiantong Jiang, Zeyi Wen, Atif Mansoor, Ajmal Mian

Research output: Chapter in Book/Conference paper › Other chapter contribution › peer-review

3 Citations (Scopus)

Abstract

Bayesian networks (BNs) are attractive, because they are graphical and interpretable machine learning models. However, exact inference on BNs is time-consuming, especially for complex problems. To improve the efficiency, we propose a fast BN exact inference solution named Fast-BNI on multi-core CPUs. Fast-BNI enhances the efficiency of exact inference through hybrid parallelism that tightly integrates coarse- and fine-grained parallelism. We also propose techniques to further simplify the bottleneck operations of BN exact inference.

Original language	English
Title of host publication	PPoPP '23: Proceedings of the 28th ACM SIGPLAN Annual Symposium on Principles and Practice of Parallel Programming
Publisher	Association for Computing Machinery (ACM)
Pages	425-426
Number of pages	2
ISBN (Electronic)	979-8-4007-0015-6
DOIs	https://doi.org/10.1145/3572848.3577476
Publication status	Published - 25 Feb 2023
Event	28th ACM SIGPLAN Annual Symposium on Principles and Practice of Parallel Programming, PPoPP 2023 - Montreal, Canada Duration: 25 Feb 2023 → 1 Mar 2023

Conference

Conference	28th ACM SIGPLAN Annual Symposium on Principles and Practice of Parallel Programming, PPoPP 2023
Country/Territory	Canada
City	Montreal
Period	25/02/23 → 1/03/23

Access to Document

10.1145/3572848.3577476

https://arxiv.org/pdf/2212.04241.pdf

Cite this

@inbook{af7d27b7c01d42a5a11308f91e823150,

title = "POSTER: Fast Parallel Exact Inference on Bayesian Networks",

abstract = "Bayesian networks (BNs) are attractive, because they are graphical and interpretable machine learning models. However, exact inference on BNs is time-consuming, especially for complex problems. To improve the efficiency, we propose a fast BN exact inference solution named Fast-BNI on multi-core CPUs. Fast-BNI enhances the efficiency of exact inference through hybrid parallelism that tightly integrates coarse- and fine-grained parallelism. We also propose techniques to further simplify the bottleneck operations of BN exact inference.",

keywords = "bayesian networks, inference, junction tree",

author = "Jiantong Jiang and Zeyi Wen and Atif Mansoor and Ajmal Mian",

note = "Funding Information: Professor Ajmal Mian is the recipient of an Australian Research Council Future Fellowship Award (project number FT210100268) funded by the Australian Government. Zeyi Wen is the corresponding author. Publisher Copyright: {\textcopyright} 2023 Owner/Author.; 28th ACM SIGPLAN Annual Symposium on Principles and Practice of Parallel Programming, PPoPP 2023 ; Conference date: 25-02-2023 Through 01-03-2023",

year = "2023",

month = feb,

day = "25",

doi = "10.1145/3572848.3577476",

language = "English",

pages = "425--426",

booktitle = "PPoPP '23: Proceedings of the 28th ACM SIGPLAN Annual Symposium on Principles and Practice of Parallel Programming",

publisher = "Association for Computing Machinery (ACM)",

address = "United States",

}

Jiang, J, Wen, Z, Mansoor, A & Mian, A 2023, POSTER: Fast Parallel Exact Inference on Bayesian Networks. in PPoPP '23: Proceedings of the 28th ACM SIGPLAN Annual Symposium on Principles and Practice of Parallel Programming. Association for Computing Machinery (ACM), pp. 425-426, 28th ACM SIGPLAN Annual Symposium on Principles and Practice of Parallel Programming, PPoPP 2023, Montreal, Canada, 25/02/23. https://doi.org/10.1145/3572848.3577476

POSTER: Fast Parallel Exact Inference on Bayesian Networks. / Jiang, Jiantong; Wen, Zeyi; Mansoor, Atif et al.
PPoPP '23: Proceedings of the 28th ACM SIGPLAN Annual Symposium on Principles and Practice of Parallel Programming. Association for Computing Machinery (ACM), 2023. p. 425-426.

Research output: Chapter in Book/Conference paper › Other chapter contribution › peer-review

TY - CHAP

T1 - POSTER

T2 - 28th ACM SIGPLAN Annual Symposium on Principles and Practice of Parallel Programming, PPoPP 2023

AU - Jiang, Jiantong

AU - Wen, Zeyi

AU - Mansoor, Atif

AU - Mian, Ajmal

N1 - Funding Information: Professor Ajmal Mian is the recipient of an Australian Research Council Future Fellowship Award (project number FT210100268) funded by the Australian Government. Zeyi Wen is the corresponding author. Publisher Copyright: © 2023 Owner/Author.

PY - 2023/2/25

Y1 - 2023/2/25

N2 - Bayesian networks (BNs) are attractive, because they are graphical and interpretable machine learning models. However, exact inference on BNs is time-consuming, especially for complex problems. To improve the efficiency, we propose a fast BN exact inference solution named Fast-BNI on multi-core CPUs. Fast-BNI enhances the efficiency of exact inference through hybrid parallelism that tightly integrates coarse- and fine-grained parallelism. We also propose techniques to further simplify the bottleneck operations of BN exact inference.

AB - Bayesian networks (BNs) are attractive, because they are graphical and interpretable machine learning models. However, exact inference on BNs is time-consuming, especially for complex problems. To improve the efficiency, we propose a fast BN exact inference solution named Fast-BNI on multi-core CPUs. Fast-BNI enhances the efficiency of exact inference through hybrid parallelism that tightly integrates coarse- and fine-grained parallelism. We also propose techniques to further simplify the bottleneck operations of BN exact inference.

KW - bayesian networks

KW - inference

KW - junction tree

UR - http://www.scopus.com/inward/record.url?scp=85149282447&partnerID=8YFLogxK

UR - https://github.com/jjiantong/FastBN.

U2 - 10.1145/3572848.3577476

DO - 10.1145/3572848.3577476

M3 - Other chapter contribution

AN - SCOPUS:85149282447

SP - 425

EP - 426

BT - PPoPP '23: Proceedings of the 28th ACM SIGPLAN Annual Symposium on Principles and Practice of Parallel Programming

PB - Association for Computing Machinery (ACM)

Y2 - 25 February 2023 through 1 March 2023

ER -

POSTER: Fast Parallel Exact Inference on Bayesian Networks

Abstract

Conference

Access to Document

Other files and links

Fingerprint

Cite this