A Component-Wise EM Algorithm for Mixtures
Résumé
In some situations, EM algorithm shows slow convergence problems. One possible reason is that standard procedures update the parameters simultaneous- ly. In this paper we focus on finite mixture estimation. In this framework, we propose a component-wise EM, which updates the parameters sequentially. We give an interpretation of this procedure as a proximal point algorithm and use it to prove the convergence. Illustrative numerical experiments show how our algorithm compares to EM and a version of the SAGE algorithm.