Kernel functions are often cited as a mechanism to encode prior knowledge of a learning task. But it can be difficult to capture prior knowledge effectively. For example, we know that image pixels of a handwritten character result from a few strokes from a single writing implement; it is not clear how to express this in a kernel function. We investigate an explanation based learning (EBL) paradigm to generate specialized kernel functions. These embody novel high-level features that are automatically constructed from the interaction of prior knowledge and training examples. Our empirical results showed that the performance of the resulting SVM surpasses that of a conventional SVM on the challenging task of classifying handwritten Chinese characters.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Feature kernel functions: improving SVMs using high-level knowledge


    Contributors:
    Sun, Q. (author) / DeJong, G. (author)


    Publication date :

    2005-01-01


    Size :

    173125 byte





    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English




    Part level transfer regularization for enhancing exemplar SVMs

    Aytar, Y. / Zisserman, A. | British Library Online Contents | 2015


    Space Vision Marker System (SVMS)

    Bondy, Michel / Krishnasamy, Rubakumar / Crymble, Derry et al. | AIAA | 2007


    A Complementary SVMS-Based Image Annotation System

    Han, Y. / Qi, X. | British Library Conference Proceedings | 2005


    Learning Interpretable SVMs for Biological Sequence Classification

    Sonnenburg, S. / Ratsch, G. / Schafer, C. | British Library Conference Proceedings | 2005