概率图模型学习技术研究进展

被引：22

作者：

刘建伟

黎海恩

罗雄麟

机构：

[1] 中国石油大学(北京)自动化研究所

来源：

自动化学报 | 2014年 / 40卷 / 06期

关键词：

概率图模型; 贝叶斯网络; 马尔科夫网络; 参数学习; 结构学习; 不完备数据集;

D O I：

暂无

中图分类号：

TP181 [自动推理、机器学习];

学科分类号：

摘要：

概率图模型能有效处理不确定性推理,从样本数据中准确高效地学习概率图模型是其在实际应用中的关键问题.概率图模型的表示由参数和结构两部分组成,其学习算法也相应分为参数学习与结构学习.本文详细介绍了基于概率图模型网络的参数学习与结构学习算法,并根据数据集是否完备而分别讨论各种情况下的参数学习算法,还针对结构学习算法特点的不同把结构学习算法归纳为基于约束的学习、基于评分搜索的学习、混合学习、动态规划结构学习、模型平均结构学习和不完备数据集的结构学习.并总结了马尔科夫网络的参数学习与结构学习算法.最后指出了概率图模型学习的开放性问题以及进一步的研究方向.

引用

页码：1025 / 1044

页数：20

共 41 条

[1] Efficient methods for learning Bayesian network super-structures[J] . Edwin Villanueva,Carlos Dias Maciel.Neurocomputing . 2013
[2] A review on evolutionary algorithms in Bayesian network learning and inference tasks[J] . Pedro Larra?aga,Hossein Karshenas,Concha Bielza,Roberto Santana.Information Sciences . 2013
[3] An artificial bee colony algorithm for learning Bayesian networks
Ji, Junzhong
Wei, Hongkai
Liu, Chunnian
[J]. SOFT COMPUTING, 2013, 17 (06) : 983 - 994
[4] Characteristic imsets for learning Bayesian network structure
Hemmecke, Raymond
Lindner, Silvia
Studeny, Milan
[J]. INTERNATIONAL JOURNAL OF APPROXIMATE REASONING, 2012, 53 (09) : 1336 - 1349
[5] Bilateral Markov mesh random field and its application to image restoration
Yousefi, S.
Kehtarnavaz, N.
Cao, Y.
Razlighi, Q. R.
[J]. JOURNAL OF VISUAL COMMUNICATION AND IMAGE REPRESENTATION, 2012, 23 (07) : 1051 - 1059
[6] Improved algorithm based on mutual information for learning Bayesian network structures in the space of equivalence classes
Li, Bing Han
Liu, San Yang
Li, Zhan Guo
[J]. MULTIMEDIA TOOLS AND APPLICATIONS, 2012, 60 (01) : 129 - 137
[7] Overview on Bayesian networks applications for dependability, risk analysis and maintenance areas[J] . P. Weber,G. Medina-Oliva,C. Simon,B. Iung.Engineering Applications of Artificial Intelligence . 2010 (4)
[8] Preserving objects in Markov Random Fields region growing image segmentation
Dawoud, Amer
Netchaev, Anton
[J]. PATTERN ANALYSIS AND APPLICATIONS, 2012, 15 (02) : 155 - 161
[9] Learning locally minimax optimal Bayesian networks
Silander, Tomi
Roos, Teemu
Myllymaki, Petri
[J]. INTERNATIONAL JOURNAL OF APPROXIMATE REASONING, 2010, 51 (05) : 544 - 557
[10] A geometric view on learning Bayesian network structures
Studeny, Milan
Vomlel, Jiri
Hemmecke, Raymond
[J]. INTERNATIONAL JOURNAL OF APPROXIMATE REASONING, 2010, 51 (05) : 573 - 586

← 1 2 3 4 5 →