Properties of the Stochastic Approximation EM Algorithm with Mini-batch Sampling - Archive ouverte HAL Access content directly
Journal Articles Statistics and Computing Year : 2020

Properties of the Stochastic Approximation EM Algorithm with Mini-batch Sampling

Abstract

To deal with very large datasets a mini-batch version of the Monte Carlo Markov Chain Stochastic Approximation Expectation– Maximization algorithm for general latent variable models is proposed. For exponential models the algorithm is shown to be convergent under classical conditions as the number of iterations increases. Numerical experiments illustrate the performance of the mini-batch algorithm in various models. In particular, we highlight that mini-batch sampling results in an important speed-up of the convergence of the sequence of estimators generated by the algorithm. Moreover, insights on the effect of the mini-batch size on the limit distribution are presented. Finally, we illustrate how to use mini-batch sampling in practice to improve results when a constraint on the computing time is given.
Fichier principal
Vignette du fichier
preprint.pdf (1.01 Mo) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-02189215 , version 1 (19-07-2019)
hal-02189215 , version 2 (18-12-2019)
hal-02189215 , version 3 (30-04-2020)

Identifiers

Cite

Estelle Kuhn, Catherine Matias, Tabea Rebafka. Properties of the Stochastic Approximation EM Algorithm with Mini-batch Sampling. Statistics and Computing, 2020, 30 (6), pp.1725-1739. ⟨hal-02189215v1⟩
236 View
245 Download

Altmetric

Share

Gmail Facebook Twitter LinkedIn More