Skip to Main content Skip to Navigation
Conference papers

Genetic Programming for Kernel-based Learning with Co-evolving Subsets Selection

Christian Gagné 1, 2 Marc Schoenauer 1, 3 Michèle Sebag 3 Marco Tomassini 2
1 TANC - Algorithmic number theory for cryptology
Inria Saclay - Ile de France, LIX - Laboratoire d'informatique de l'École polytechnique [Palaiseau]
Abstract : Support Vector Machines (SVMs) are well-established Machine Learning (ML) algorithms. They rely on the fact that i) linear learning can be formalized as a well-posed optimization problem; ii) non-linear 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.
Document type :
Conference papers
Complete list of metadatas

Cited literature [18 references]  Display  Hide  Download

https://hal.inria.fr/inria-00116344
Contributor : Marc Schoenauer <>
Submitted on : Sunday, November 26, 2006 - 11:45:19 AM
Last modification on : Wednesday, September 16, 2020 - 4:52:16 PM
Long-term archiving on: : Tuesday, April 6, 2010 - 11:24:31 PM

Files

evokern-paper.pdf
Files produced by the author(s)

Identifiers

Collections

Citation

Christian Gagné, Marc Schoenauer, Michèle Sebag, Marco Tomassini. Genetic Programming for Kernel-based Learning with Co-evolving Subsets Selection. PPSN'06, Sep 2006, Reykjavik, pp.1008-1017. ⟨inria-00116344⟩

Share

Metrics

Record views

449

Files downloads

1106