The convergence rate of the MDM algorithm

In this paper we will describe a simple proof of a linear convergence rate for the MDM algorithm that solves the Minimum Norm Problem (MNP). Linear convergence rates have been shown for the SMO algorithm, but the proofs require specific assumptions and are rather involved. We will follow a different approach, with a more geometric flavor. While as of now our proof also requires some of the just mentioned assumptions, we shall discuss some examples where linear convergence holds without them, suggesting that a linear convergence rate may be achieved under conditions more general than those currently known.