The max-min hill-climbing Bayesian network structure learning algorithm

被引:1268
作者
Tsamardinos, Ioannis [1 ]
Brown, Laura E. [1 ]
Aliferis, Constantin F. [1 ]
机构
[1] Vanderbilt Univ, Dept Biomed Informat, Discovery Syst Lab, Nashville, TN 37232 USA
关键词
Bayesian networks; graphical models; structure learning;
D O I
10.1007/s10994-006-6889-7
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
We present a new algorithm for Bayesian network structure learning, called Max-Min Hill-Climbing (MMHC). The algorithm combines ideas from local learning, constraint-based, and search-and-score techniques in a principled and effective way. It first reconstructs the skeleton of a Bayesian network and then performs a Bayesian-scoring greedy hill-climbing search to orient the edges. In our extensive empirical evaluation MMHC outperforms on average and in terms of various metrics several prototypical and state-of-the-art algorithms, namely the PC, Sparse Candidate, Three Phase Dependency Analysis, Optimal Reinsertion, Greedy Equivalence Search, and Greedy Search. These are the first empirical results simultaneously comparing most of the major Bayesian network algorithms against each other. MMHC offers certain theoretical advantages, specifically over the Sparse Candidate algorithm, corroborated by our experiments. MMHC and detailed results of our study are publicly available at http://www.dsl-lab.org/supplements/mmhc_paper/mmhc_index.html.
引用
收藏
页码:31 / 78
页数:48
相关论文
共 81 条
  • [1] Hailfinder: A Bayesian system for forecasting severe weather
    Abramson, B
    Brown, J
    Edwards, W
    Murphy, A
    Winkler, RL
    [J]. INTERNATIONAL JOURNAL OF FORECASTING, 1996, 12 (01) : 57 - 71
  • [2] A comparison of learning algorithms for Bayesian networks:: a case study based on data from an emergency medical service
    Acid, S
    de Campos, LM
    Fernández-Luna, JM
    Rodríguez, S
    Rodríguez, JM
    Salcedo, JL
    [J]. ARTIFICIAL INTELLIGENCE IN MEDICINE, 2004, 30 (03) : 215 - 232
  • [3] Searching for Bayesian network structures in the space of restricted acyclic partially directed graphs
    Acid, S
    de Campos, LM
    [J]. JOURNAL OF ARTIFICIAL INTELLIGENCE RESEARCH, 2003, 18 : 445 - 490
  • [4] ACID S, 2001, INT J APPROX REASON, P235
  • [5] NEW LOOK AT STATISTICAL-MODEL IDENTIFICATION
    AKAIKE, H
    [J]. IEEE TRANSACTIONS ON AUTOMATIC CONTROL, 1974, AC19 (06) : 716 - 723
  • [6] Aliferis C F, 2003, AMIA Annu Symp Proc, P21
  • [7] Aliferis CF, 2003, METMBS'03: PROCEEDINGS OF THE INTERNATIONAL CONFERENCE ON MATHEMATICS AND ENGINEERING TECHNIQUES IN MEDICINE AND BIOLOGICAL SCIENCES, P371
  • [8] Andreassen S., 1989, COMPUTER AIDED ELECT
  • [9] [Anonymous], 1998, LEARNING BAYESIAN NE
  • [10] [Anonymous], 1999, Probabilistic Networks and Expert Systems