Expectation-maximization algorithms, or EM algorithms for short, are iterative algorithms designed to solve maximum likelihood estimation problems. The general setting is that one observes a random sample Y1; Y2;:::; Yn of a random variable Y whose probability density function (pdf) f. (· | x0) with respect to some (known) dominating measure is known up to an unknown “parameter” x0. The goal is to estimate x0 and, one might add, to do it well. In this chapter, that means to solve the maximum likelihood problem.
CITATION STYLE
Byrne, C., & Eggermont, P. P. B. (2015). Em algorithms. In Handbook of Mathematical Methods in Imaging: Volume 1, Second Edition (pp. 305–388). Springer New York. https://doi.org/10.1007/978-1-4939-0790-8_8
Mendeley helps you to discover research relevant for your work.