- author = "Christian Gagne and Marc Schoenauer and Michele Sebag and Marco Tomassini",
- title = "Genetic Programming for Kernel-Based Learning with Co-evolving Subsets Selection",
- booktitle = "Parallel Problem Solving from Nature - PPSN IX",
- year = "2006",
- editor = "Thomas Philip Runarsson and Hans-Georg Beyer and Edmund Burke and Juan J. Merelo-Guervos and L. Darrell Whitley and Xin Yao",
- volume = "4193",
- pages = "1008--1017",
- series = "LNCS",
- address = "Reykjavik, Iceland",
- publisher_address = "Berlin",
- month = "9-13 " # sep,
- publisher = "Springer-Verlag",
- ISBN = "3-540-38990-3",
- keywords = "genetic algorithms, genetic programming, hyperheuristic, DSS, coevolution, open beagle",
-
URL = "
http://ppsn2006.raunvis.hi.is/proceedings/287.pdf",
-
URL = "
http://arxiv.org/abs/cs/0611135",
-
DOI = "
doi:10.1007/11844297_102",
- size = "10 pages",
- abstract = "Support Vector Machines (SVMs) are well-established Machine Learning (ML) algorithms. They rely on the fact that i) linear learning can be formalised as a well-posed optimisation problem; ii) nonlinear learning can be brought into linear learning thanks to the kernel trick and the mapping of the initial search space onto a high dimensional feature space. The kernel is designed by the ML expert and it governs the efficiency of the SVM approach. In this paper, a new approach for the automatic design of kernels by Genetic Programming, called the Evolutionary Kernel Machine (EKM), is presented. EKM combines a well-founded fitness function inspired from the margin criterion, and a co-evolution framework ensuring the computational scalability of the approach. Empirical validation on standard ML benchmark demonstrates that EKM is competitive using state-of-the-art SVMs with tuned hyper-parameters.",
-
notes = "PPSN-IX
evolved Kernels are forced to be symmetric functions. Mercer's condition not enforced, but evolved. 3 co-evolving populations. runtime < 1 hour. Size based parsimony pressure. Comparison with k-nn nearest neighbours and SVM, GK-SVM (both with somewhat optimised parameters). 6 undemanding UCI benchmarks.",
