Evolving kernel functions for SVMs by genetic programming

Created by W.Langdon from gp-bibliography.bib Revision:1.4216

  title =        "Evolving kernel functions for SVMs by genetic
  author =       "Laura Diosan and Alexandrina Rogozan and 
                 Jean-Pierre Pecuchet",
  booktitle =    "Sixth International Conference on Machine Learning and
                 Applications, ICMLA 2007",
  year =         "2007",
  month =        "13-15 " # dec,
  pages =        "19--24",
  address =      "Cincinnati, Ohio, USA",
  publisher =    "IEEE",
  keywords =     "genetic algorithms, genetic programming, support
                 vector machines, GP chromosome, SVM kernel functions,
                 evolved kernel, kernel expression, mathematical
                 expression, tree encoding",
  DOI =          "doi:10.1109/ICMLA.2007.70",
  abstract =     "hybrid model for evolving support vector machine (SVM)
                 kernel functions is developed in this paper. The kernel
                 expression is considered as a parameter of the SVM
                 algorithm and the current approach tries to find the
                 best expression for this SVM parameter. The model is a
                 hybrid technique that combines a genetic programming
                 (GP) algorithm and a support vector machine (SVM)
                 algorithm. Each GP chromosome is a tree encoding the
                 mathematical expression for the kernel function. The
                 evolved kernel is compared to several human-designed
                 kernels and to a previous genetic kernel on several
                 datasets. Numerical experiments show that the SVM
                 embedding our evolved kernel performs statistically
                 better than standard kernels, but also than previous
                 genetic kernel for all considered classification
  notes =        "also known as \cite{4457202}.

Genetic Programming entries for Laura Diosan Alexandrina Rogozan Jean Pierre Pecuchet