trons have a dual kernel representation implementing
the same decision function The optimal margin algorithm exploits this duality both for improved eciency
and exibility In the dual space the decision function
is expressed as a linear combination of basis functions
parametrized by the supporting patterns The support
ing patterns correspond to the class centers of RBF
classiers and are chosen automatically by the maxi
mum margin training procedure In the case of polyno
mial classiers