Part of Advances in Neural Information Processing Systems 11 (NIPS 1998)
John Platt
Training a Support Vector Machine (SVM) requires the solution of a very large quadratic programming (QP) problem. This paper proposes an al(cid:173) gorithm for training SVMs: Sequential Minimal Optimization, or SMO. SMO breaks the large QP problem into a series of smallest possible QP problems which are analytically solvable. Thus, SMO does not require a numerical QP library. SMO's computation time is dominated by eval(cid:173) uation of the kernel, hence kernel optimizations substantially quicken SMO. For the MNIST database, SMO is 1.7 times as fast as PCG chunk(cid:173) ing; while for the UCI Adult database and linear SVMs, SMO can be 1500 times faster than the PCG chunking algorithm.