Classification on Proximity Data with LP- and QP-Machines
Mr Ralf Herbrich, Department of Computer Science, Technical University Berlin

Abstract: We provide a new linear program to deal with classification of data in the case of functions written in terms of pairwise proximities. This allows us to avoid the problems inherent in using feature spaces with indefinite metric in Support Vector Machines, since the notion of a margin is purely needed in input space where the classification actually occurs. Moreover in our approach we can enforce sparsity in the proximity representation by sacrificing training error. This turns out to be favorable for proximity data. Similar to \nu-SV methods, the only parameter needed in the algorithm is the (asymptotical) number of data points being classified with a margin. Finally, the algorithm is successfully compared with \nu-SV learning in proximity space and K-nearest-neighbors on real world data from Neuroscience and molecular biology. The presented work is joint work with Alex Smola, Thore Graepel, and Bernhard Schölkopf.

This seminar was held on 10 March 1999 at the Department of Computer Science, Royal Holloway, University of London

back