ABSTRACT
This chapter gives an overview of the implementation of Support Vector Machines. We take Algorithm 4.3.1 (C-support vector classification) as the representative. Suppose the training set is given by
T = {(x1, y1), · · · , (xl, yl)}, (7.0.1)
where xi ∈ Rn, yi ∈ Y = {1,−1}, i = 1, · · · , l, now the convex quadratic programming problem (4.3.1)∼(4.3.3) w.r.t. variable α = (α1, · · · , αl)T
min α
yiyjαiαjK(xi, xj)− l∑ i=1
αi , (7.0.2)
s.t.