Mean field theory for sigmoid belief networks

被引：202

作者：

Saul, LK

Jaakkola, T

Jordan, MI

机构：

[1] Ctr. for Biol./Compl. Learning, Massachusetts Inst. of Technology, Cambridge, MA 02139, 79 Amherst Street

来源：

JOURNAL OF ARTIFICIAL INTELLIGENCE RESEARCH | 1996年 / 4卷

关键词：

D O I：

10.1613/jair.251

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

We develop a mean field theory for sigmoid belief networks based on ideas from statistical mechanics. Our mean field theory provides a tractable approximation to the true probability distribution in these networks; it also yields a lower bound on the likelihood of evidence. We demonstrate the utility of this framework on a benchmark problem in statistical pattern recognition-the classification of handwritten digits.

引用

页码：61 / 76

页数：16

共 50 条

[1] Mean field theory for sigmoid belief networks
Saul, Lawrence K.
Jaakkola, Tommi
Jordan, Michael I.
Journal of Artificial Intelligence Research, 1996, 4 : 61 - 76
[2] A variational mean-field theory for sigmoidal belief networks
Bhattacharyya, C
Keerthi, SS
ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 13, 2001, 13 : 374 - 380
[3] A mean field approach to MAP in belief networks
Peng, Y
Jin, M
IJCNN 2000: PROCEEDINGS OF THE IEEE-INNS-ENNS INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS, VOL V, 2000, : 652 - 657
[4] Mean-field methods for a special class of Belief Networks
Bhattacharyya, C
Keerthi, SS
JOURNAL OF ARTIFICIAL INTELLIGENCE RESEARCH, 2001, 15 : 91 - 114
[5] Mean-field methods for a special class of Belief networks
Dept. of Computer Science and Automation, Indian Institute of Science, Bangalore, Karnataka, 560012, India
不详
J Artif Intell Res, (91-114):
[6] Beyond mean field theory: statistical field theory for neural networks
Buicel, Michael A.
Chow, Carson C.
JOURNAL OF STATISTICAL MECHANICS-THEORY AND EXPERIMENT, 2013,
[7] Incremental sigmoid belief networks for grammar learning
Department of Computer Science, University of Geneva, 7 route de Drize, A 1227, Carouge, Switzerland
不详
J. Mach. Learn. Res., (3541-3570):
[8] Incremental Sigmoid Belief Networks for Grammar Learning
Henderson, James
Titov, Ivan
JOURNAL OF MACHINE LEARNING RESEARCH, 2010, 11 : 3541 - 3570
[9] Mean field theory for asymmetric neural networks
Kappen, HJ
Spanjers, JJ
PHYSICAL REVIEW E, 2000, 61 (05): : 5658 - 5663
[10] Mean field theory for asymmetric neural networks
Kappen, H.J.
Spanjers, J.J.
Physical Review E - Statistical, Nonlinear, and Soft Matter Physics, 2000, 61 (05): : 5658 - 5663

← 1 2 3 4 5 →