The paper introduces Voting EM, an adaptive online learning algorithm of Bayesian network parameters. Voting EM is an extension of the EM( ) algorithm suggested by [1]. We show convergence properties of the Voting EM that uses a constant learning rate. We use the convergence properties to formulate an error driven scheme for adapting the learning rate. The resultant algorithm converges with the...

In this paper, we present a new parallel EM algorithm with the optimal data replication on a hypercube niultiprocessor. Although data replication has been frequently used to reduce data sharing overhead, none of the parallel EM algorithms have attempted to optimize the data replication. To maximize efficiency of the proposed parallel EM algorithm, data replication have been optimized through st...

Information geometry of partial likelihood is constructed and is used to derive the em-algorithm for learning parameters of a conditional distribution model through information -theoretic projections. To construct the coordinates of the information geometry, an Expectation-Maximization (EM) framework is described for the distribution learning problem using the Gaussian mixture probability model...

Like any other precipitation process, in supercritical water hydrothermal synthesis (SWHS), the need to improve product quality and minimize production cost requires understanding and optimization of Particle Size Distribution (PSD). In this work, using Population Balance Equation (PBE) containing nucleation and growth terms, the reactive precipitation of zirconia nanoparticles prepared by ...

The expectation maximization (EM) algorithm is popular in estimating the parameters of the statistical models. In this paper, we consider application of the EM algorithm to Maximum Likelihood estimation. A Hidden Markov Model (HMM) formulation is used and EM algorithm is applied to estimate the parameters of the HMM which, in turn, are used to estimate received amplitudes of the users. The prop...

Mixture probability densities are popular models that are used in several data mining and machine learning applications, e.g., clustering. A standard algorithm for learning such models from data is the Expectation-Maximization (EM) algorithm. However, EM can be slow with large datasets, and therefore approximation techniques are needed. In this paper we propose a variational approximation to th...

in this paper, we consider the fuzzy fixed-charge transportation problem (ffctp). both of fixed and transportation cost are fuzzy numbers. contrary to previous works, electromagnetism-like algorithms (em) is firstly proposed in this research area to solve the problem. three types of em; original em, revised em, and hybrid em are firstly employed for the given problem. the latter is being firstl...

The EM algorithm is a popular and useful algorithm for "nding the maximum likelihood estimator in incomplete data problems. Each iteration of the algorithm consists of two simple steps: an E-step, in which a conditional expectation is calculated, and an M-step, where the expectation is maximized. In some problems, however, the EM algorithm cannot be applied since the conditional expectation req...

4 The Expectation-Maximization algorithm 7 4.1 Jointly-non-concave incomplete log-likelihood . . . . . . . . . . . 7 4.2 (Possibly) Concave complete data log-likelihood . . . . . . . . . . 8 4.3 The general EM derivation . . . . . . . . . . . . . . . . . . . . . 9 4.4 The E& M-steps . . . . . . . . . . . . . . . . . . . . . . . . . . 11 4.5 The EM algorithm . . . . . . . . . . . . . . . . . . ....

The approach proposed in this paper takes into account the uncertainty in colour modelling by employing variational Bayesian estimation. Mixtures of Gaussians are considered for modelling colour images. Distributions of parameters characterising colour regions are inferred from data statistics. The Variational Expectation-Maximization (VEM) algorithm is used for estimating the hyperparameters c...