The multiset EM algorithm

TL;DR: In this paper, a group of Channel Matching (CM) algorithms are developed for machine learning, which can outperform the EM algorithm when the mixture ratios are imbalanced, or when local convergence exists.

TL;DR: It is proved that the relative entropy between the sampling distribution and predicted distribution may be equal to R − G, hence, solving the maximum likelihood mixture model only needs minimizing R −G, without needing Jensen’s inequality.

TL;DR: It is concluded that the optimal value of filling missing data is found by the algorithm of this paper to speed up the convergence rate, strengthened the stability of clustering, data filling effect is remarkable.

TL;DR: In this article , the authors introduced the theory of multilevel hierarchical data Expectation Maximization (EM)-algorithm via discrete time Markov chain (DTMC) epidemic models.

TL;DR: If data augmentation can be used in the calculation of the maximum likelihood estimate, then in the same cases one ought to be able to use it in the computation of the posterior distribution of parameters of interest.

TL;DR: In many cases, complete-data maximum likelihood estimation is relatively simple when conditional on some function of the parameters being estimated as mentioned in this paper, and convergence is stable, with each iteration increasing the likelihood.

TL;DR: Two modifications to the MCEM algorithm (the poor man's data augmentation algorithms), which allow for the calculation of the entire posterior, are presented and serve as diagnostics for the validity of the posterior distribution.

TL;DR: This parameter-expanded Ei M, PX-EM, algorithm shares the simplicity and stability of ordinary EM, but has a faster rate of convergence since its M step performs a more efficient analysis.