Universal learning curves of support vector machines
Opper, Manfred and Urbanczik, Robert (2001) Universal learning curves of support vector machines. Physical Review Letters, 86, 4410-4413.
Using methods of Statistical Physics, we investigate the role of model complexity in learning with support vector machines (SVMs), which are an important alternative to neural networks. We show the advantages of using SVMs with kernels of infinite complexity on noisy target rules, which, in contrast to common theoretical beliefs, are found to achieve optimal generalization error although the training error does not converge to the generalization error. Moreover, we find a universal asymptotics of the learning curves which only depend on the target rule but not on the SVM kernel.
|Divisions:||Faculty of Physical and Applied Science > Electronics and Computer Science
|Date Deposited:||15 Mar 2004|
|Last Modified:||15 Aug 2012 03:12|
|Contributors:||Opper, Manfred (Author)
Urbanczik, Robert (Author)
|Further Information:||Google Scholar|
|ISI Citation Count:||18|
|RDF:||RDF+N-Triples, RDF+N3, RDF+XML, Browse.|
Actions (login required)