Title :
A Theoretical and Experimental Comparison of the EM and SEM Algorithm
Author :
Blomer, J. ; Bujna, K. ; Kuntze, D.
Author_Institution :
Dept. of Comput. Sci., Univ. of Paderborn, Paderborn, Germany
Abstract :
In this paper we provide a new analysis of the SEM algorithm. Unlike previous work, we focus on the analysis of a single run of the algorithm. First, we discuss the algorithm for general mixture distributions. Second, we consider Gaussian mixture models and show that with high probability the update equations of the EM algorithm and its stochastic variant are almost the same, given that the input set is sufficiently large. Our experiments confirm that this still holds for a large number of successive update steps. In particular, for Gaussian mixture models, we show that the stochastic variant runs nearly twice as fast.
Keywords :
Gaussian processes; data handling; expectation-maximisation algorithm; mixture models; statistical distributions; Gaussian mixture models; SEM algorithm; expectation-maximization algorithm; general mixture distributions; stochastic variant; Algorithm design and analysis; Covariance matrices; Gaussian mixture model; Numerical analysis; Vectors;
Conference_Titel :
Pattern Recognition (ICPR), 2014 22nd International Conference on
Conference_Location :
Stockholm
DOI :
10.1109/ICPR.2014.253