A New Class of EM Algorithms. Escaping Local Minima and Handling Intractable Sampling - Archive ouverte HAL Access content directly
Preprints, Working Papers, ... Year :

A New Class of EM Algorithms. Escaping Local Minima and Handling Intractable Sampling

(1) , (2)
1
2

Abstract

The expectation-maximization (EM) algorithm is a powerful computational technique for maximum likelihood estimation in incomplete data models. When the expectation step cannot be performed in closed form, a stochastic approximation of EM (SAEM) can be used. The convergence of the SAEM toward local maxima of the observed likelihood has been proved and its numerical efficiency has been demonstrated. However, despite appealing features, the limit position of this algorithm can strongly depend on its starting position. Moreover, sampling from the posterior distribution may be intractable or have a high computational cost. To cope with this two issues, we propose here a new stochastic approximation version of the EM in which we do not sample from the exact distribution in the expectation phase of the procedure. We first prove the convergence of this algorithm toward local maxima of the observed likelihood. Then, we propose an instantiation of this general procedure to favor convergence toward global maxima. Experiments on synthetic and real data highlight the performance of this algorithm in comparison to the SAEM.
Fichier principal
Vignette du fichier
2019_Allassonniere_Chevallier.pdf (34.38 Mo) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-02044722 , version 1 (21-02-2019)
hal-02044722 , version 2 (26-02-2019)
hal-02044722 , version 3 (18-06-2019)
hal-02044722 , version 4 (23-04-2020)

Identifiers

  • HAL Id : hal-02044722 , version 3

Cite

Stéphanie Allassonnière, Juliette Chevallier. A New Class of EM Algorithms. Escaping Local Minima and Handling Intractable Sampling. 2019. ⟨hal-02044722v3⟩

Collections

UNIV-PARIS7 USPC
843 View
475 Download

Share

Gmail Facebook Twitter LinkedIn More