Catherine Matias (LPSM Sorbonne Université, Université de Paris, CNRS)

will speak on

Properties of the stochastic approximation EM algorithm with mini-batch sampling

Time: 12:00PM
Date: Mon 7th December 2020
Location: Online [map]

Abstract: To deal with very large data sets 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.
https://link.springer.com/article/10.1007/s11222-020-09968-0

(This talk is part of the Statistics and Actuarial Science series.)

PDF notice

Return to all seminars