Some properties of adding a smoothing step to the EM algorithm (Q583872)

The EM algorithm is an iterative statistical and numerical method that in many cases is useful for finding the maximum of a likelihood. Adding a smoothing step after the usual expectation and maximization steps of the EM algorithm results in a useful method for solving statistical problems in image analysis and integral equations. The author gives some connections between this algorithm, known as EMS, and maximizing a penalized likelihood and derives an upper bound on the convergence rate.