Semisupervised Learning Using Bayesian Interpretation: Application to LS-SVM

被引：34

作者：

Adankon, Mathias M. ^{[1
]}

Cheriet, Mohamed ^{[1
]}

Biem, Alain ^{[2
]}

机构：

[1] Univ Quebec, Ecole Technol Super, Synchromedia Lab Multimedia Commun Telepresence, Montreal, PQ H3C 1K3, Canada

[2] IBM Corp, Thomas J Watson Res Ctr, New York, NY 10598 USA

来源：

IEEE TRANSACTIONS ON NEURAL NETWORKS | 2011年 / 22卷 / 04期

基金：

加拿大自然科学与工程研究理事会;

关键词：

Bayesian inference; kernel machine; least-square support vector machine (SVM); semisupervised learning; SVM; OUT CROSS-VALIDATION; VECTOR MACHINES; MODEL SELECTION;

D O I：

10.1109/TNN.2011.2105888

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Bayesian reasoning provides an ideal basis for representing and manipulating uncertain knowledge, with the result that many interesting algorithms in machine learning are based on Bayesian inference. In this paper, we use the Bayesian approach with one and two levels of inference to model the semisupervised learning problem and give its application to the successful kernel classifier support vector machine (SVM) and its variant least-squares SVM (LS-SVM). Taking advantage of Bayesian interpretation of LS-SVM, we develop a semisupervised learning algorithm for Bayesian LS-SVM using our approach based on two levels of inference. Experimental results on both artificial and real pattern recognition problems show the utility of our method.

引用

页码：513 / 524

页数：12

共 50 条

[1] Genetic algorithm-based training for semi-supervised SVM [J].