The expectation-maximization algorithm alternates an E-step, which takes the conditional expectation of a complete-data log objective over latent variables, and an M-step, which maximizes that expected objective over the parameters.
Articles by others on the same topic
The Expectation-Maximization (EM) algorithm is a statistical technique used for finding maximum likelihood estimates of parameters in probabilistic models, especially when the data are incomplete or have missing values. It is commonly applied in scenarios where the model depends on latent (hidden) variables, and it's particularly useful in clustering, density estimation, and other machine learning applications.