THESIS
2006
xiv, 143 leaves : ill. ; 30 cm
Abstract
Many machine learning and pattern recognition algorithms rely on a distance metric. Instead of choosing a metric manually, a more promising approach is to learn the metric from data automatically. Besides some early work on met-ric learning for classification, more and more efforts have been devoted in recent years to learning a distance metric under the semi-supervised learning setting. Semi-supervised learning is a learning paradigm between the supervised and un-supervised learning extremes. Algorithms of this class usually solve the classifi-cation or clustering problems with the aid of additional background knowledge. While there has been a whole set of interesting ideas on how to learn from data with supervisory information, we focus our study on semi-supervised learning in the met...[
Read more ]
Many machine learning and pattern recognition algorithms rely on a distance metric. Instead of choosing a metric manually, a more promising approach is to learn the metric from data automatically. Besides some early work on met-ric learning for classification, more and more efforts have been devoted in recent years to learning a distance metric under the semi-supervised learning setting. Semi-supervised learning is a learning paradigm between the supervised and un-supervised learning extremes. Algorithms of this class usually solve the classifi-cation or clustering problems with the aid of additional background knowledge. While there has been a whole set of interesting ideas on how to learn from data with supervisory information, we focus our study on semi-supervised learning in the metric learning context.
In this thesis, we propose a series of novel methods for semi-supervised dis-tance metric learning with additional information in the form of pairwise similar-ity and dissimilarity constraints. More specifically, metric learning in nonpara-metric and parametric forms, kernel-based metric learning, and metric learning based on manifold structure are presented in turn. We apply our methods to some real-world applications, such as content-based image retrieval and color image segmentation. Experimental results show that our proposed methods outperform previous metric learning methods.
Post a Comment