Kristian Woodsend and J. Gondzio
Abstract
A parallel implementation of Support Vector Machine training
for problems involving nonlinear kernels has been developed.
The kernel matrix is approximated by a partial Cholesky
decomposition. Theoretical issues associated with constructing
the best possible (and yet computationally efficient)
approximation are discussed in detail and implemented in OOPS.
The structure of the augmented system matrix is exploited
to partition data and computations amongst parallel processors
efficiently. The new implementation has been applied to solve
problems which involve very large data sets. Excellent parallel
efficiency was observed on such problems.
Key words: Parallel support vector machines, Interior point method, Separable quadratic program, Low-rank approximation, Cholesky factorization