Hands-On Unsupervised Learning with Python

Component selection using Bayesian Gaussian mixture

A Bayesian Gaussian mixture model is an extension of a standard Gaussian mixture based on variational framework. This topic is quite advanced, and it requires a thorough mathematical description, which is beyond the scope of this book (you can find it in Nasios N. and Bors A. G., Variational Learning for Gaussian Mixture Models, IEEE Transactions On Systems, Man, and Cybernetics, 36/ 4, 08/2006). However, before we discuss the main properties, it will be helpful to understand the main concepts and the differences. Let's suppose that we have a dataset, X, and a probabilistic model parameterized with the vector θ. In the previous sections, you saw that the probability, p(X|θ), is the likelihood, ...