Semi-Supervised Logistic Discrimination via Regularized Gaussian Basis Expansions

Abstract

The problem of constructing classification methods based on both labeled and unlabeled data sets is considered for analyzing data with complex structures. We introduce a semi-supervised logistic discriminant model with Gaussian basis expansions. Unknown parameters included in the logistic model are estimated by regularization method along with the technique of EM algorithm. For selection of adjusted parameters, we derive a model selection criterion from Bayesian viewpoints. Numerical studies are conducted to investigate the effectiveness of our proposed modeling procedures.

Keywords:

• Bayesian approach
• EM algorithm
• Logistic regression
• Regularization
• Semi-supervised learning

Mathematics Subject Classification:

• 62H30
• 62G08

Acknowledgment

The authors would like to thank the anonymous reviewer for his constructive and helpful comments.