A sparse structure learning algorithm for Gaussian Bayesian network identification from high-dimensional data

Shuai Huang, Jing Li, Jieping Ye, Adam Fleisher, Kewei Chen, Teresa Wu, Eric Reiman

Research output: Contribution to journalArticle

35 Citations (Scopus)

Abstract

Structure learning of Bayesian Networks (BNs) is an important topic in machine learning. Driven by modern applications in genetics and brain sciences, accurate and efficient learning of large-scale BN structures from high-dimensional data becomes a challenging problem. To tackle this challenge, we propose a Sparse Bayesian Network (SBN) structure learning algorithm that employs a novel formulation involving one L1-norm penalty term to impose sparsity and another penalty term to ensure that the learned BN is a Directed Acyclic Graph (DAG) - a required property of BNs. Through both theoretical analysis and extensive experiments on 11 moderate and large benchmark networks with various sample sizes, we show that SBN leads to improved learning accuracy, scalability, and efficiency as compared with 10 existing popular BN learning algorithms. We apply SBN to a real-world application of brain connectivity modeling for Alzheimer's disease (AD) and reveal findings that could lead to advancements in AD research.

Original languageEnglish (US)
Pages (from-to)1328-1342
Number of pages15
JournalIEEE Transactions on Pattern Analysis and Machine Intelligence
Volume35
Issue number6
DOIs
StatePublished - 2013

Fingerprint

Structure Learning
Bayesian networks
High-dimensional Data
Bayesian Networks
Learning algorithms
Learning Algorithm
Alzheimer's Disease
Network Structure
Penalty
Brain
L1-norm
Directed Acyclic Graph
Network Algorithms
Term
Real-world Applications
Sparsity
Learning systems
Scalability
Theoretical Analysis
Machine Learning

Keywords

  • Bayesian network
  • Data mining
  • Machine learning

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Vision and Pattern Recognition
  • Software
  • Computational Theory and Mathematics
  • Applied Mathematics

Cite this

A sparse structure learning algorithm for Gaussian Bayesian network identification from high-dimensional data. / Huang, Shuai; Li, Jing; Ye, Jieping; Fleisher, Adam; Chen, Kewei; Wu, Teresa; Reiman, Eric.

In: IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 35, No. 6, 2013, p. 1328-1342.

Research output: Contribution to journalArticle

@article{bb5aa9180a594cf7b240657d01bcde0c,
title = "A sparse structure learning algorithm for Gaussian Bayesian network identification from high-dimensional data",
abstract = "Structure learning of Bayesian Networks (BNs) is an important topic in machine learning. Driven by modern applications in genetics and brain sciences, accurate and efficient learning of large-scale BN structures from high-dimensional data becomes a challenging problem. To tackle this challenge, we propose a Sparse Bayesian Network (SBN) structure learning algorithm that employs a novel formulation involving one L1-norm penalty term to impose sparsity and another penalty term to ensure that the learned BN is a Directed Acyclic Graph (DAG) - a required property of BNs. Through both theoretical analysis and extensive experiments on 11 moderate and large benchmark networks with various sample sizes, we show that SBN leads to improved learning accuracy, scalability, and efficiency as compared with 10 existing popular BN learning algorithms. We apply SBN to a real-world application of brain connectivity modeling for Alzheimer's disease (AD) and reveal findings that could lead to advancements in AD research.",
keywords = "Bayesian network, Data mining, Machine learning",
author = "Shuai Huang and Jing Li and Jieping Ye and Adam Fleisher and Kewei Chen and Teresa Wu and Eric Reiman",
year = "2013",
doi = "10.1109/TPAMI.2012.129",
language = "English (US)",
volume = "35",
pages = "1328--1342",
journal = "IEEE Transactions on Pattern Analysis and Machine Intelligence",
issn = "0162-8828",
publisher = "IEEE Computer Society",
number = "6",

}

TY - JOUR

T1 - A sparse structure learning algorithm for Gaussian Bayesian network identification from high-dimensional data

AU - Huang, Shuai

AU - Li, Jing

AU - Ye, Jieping

AU - Fleisher, Adam

AU - Chen, Kewei

AU - Wu, Teresa

AU - Reiman, Eric

PY - 2013

Y1 - 2013

N2 - Structure learning of Bayesian Networks (BNs) is an important topic in machine learning. Driven by modern applications in genetics and brain sciences, accurate and efficient learning of large-scale BN structures from high-dimensional data becomes a challenging problem. To tackle this challenge, we propose a Sparse Bayesian Network (SBN) structure learning algorithm that employs a novel formulation involving one L1-norm penalty term to impose sparsity and another penalty term to ensure that the learned BN is a Directed Acyclic Graph (DAG) - a required property of BNs. Through both theoretical analysis and extensive experiments on 11 moderate and large benchmark networks with various sample sizes, we show that SBN leads to improved learning accuracy, scalability, and efficiency as compared with 10 existing popular BN learning algorithms. We apply SBN to a real-world application of brain connectivity modeling for Alzheimer's disease (AD) and reveal findings that could lead to advancements in AD research.

AB - Structure learning of Bayesian Networks (BNs) is an important topic in machine learning. Driven by modern applications in genetics and brain sciences, accurate and efficient learning of large-scale BN structures from high-dimensional data becomes a challenging problem. To tackle this challenge, we propose a Sparse Bayesian Network (SBN) structure learning algorithm that employs a novel formulation involving one L1-norm penalty term to impose sparsity and another penalty term to ensure that the learned BN is a Directed Acyclic Graph (DAG) - a required property of BNs. Through both theoretical analysis and extensive experiments on 11 moderate and large benchmark networks with various sample sizes, we show that SBN leads to improved learning accuracy, scalability, and efficiency as compared with 10 existing popular BN learning algorithms. We apply SBN to a real-world application of brain connectivity modeling for Alzheimer's disease (AD) and reveal findings that could lead to advancements in AD research.

KW - Bayesian network

KW - Data mining

KW - Machine learning

UR - http://www.scopus.com/inward/record.url?scp=84885013182&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84885013182&partnerID=8YFLogxK

U2 - 10.1109/TPAMI.2012.129

DO - 10.1109/TPAMI.2012.129

M3 - Article

C2 - 22665720

AN - SCOPUS:84885013182

VL - 35

SP - 1328

EP - 1342

JO - IEEE Transactions on Pattern Analysis and Machine Intelligence

JF - IEEE Transactions on Pattern Analysis and Machine Intelligence

SN - 0162-8828

IS - 6

ER -