EM algorithm and variants: an informal tutorial

TL;DR

This paper provides an informal tutorial on the EM algorithm, covering its theory and variants, highlighting the vast potential for future developments beyond current methods.

Contribution

It offers a comprehensive review of EM and its variants, emphasizing the unexplored possibilities in the field.

Findings

EM algorithm is widely applicable for maximum likelihood estimation.

Various EM variants extend its capabilities and efficiency.

There is significant potential for new EM methods beyond current research.

Abstract

The expectation-maximization (EM) algorithm introduced by Dempster et al in 1977 is a very general method to solve maximum likelihood estimation problems. In this informal report, we review the theory behind EM as well as a number of EM variants, suggesting that beyond the current state of the art is an even much wider territory still to be discovered.