An extrapolated sequential minimal optimization algorithm for support vector machines

The sequential minimal optimization (SMO) algorithm is a popular algorithm used to solve the support vector machine problem due to its efficiency and ease of implementation. We investigate applying extrapolation methods to the SMO update method in order to increase the rate of convergence of this algorithm. We first show that the update method is Newtonian and that extrapolation ensures the update is norm reducing on the objective function. We also note that choosing the working set pair according to some partial order does result in slightly faster speedups in algorithm performance.