CIAR Summer School Tutorial
Lecture 1a: Mixtures of Gaussians, EM, and Variational Free Energy
Two types of density model
(with hidden configurations h)
Some difficulties with soft k-means
A generative view of clustering
The mixture of Gaussians generative model
Computing the new mixing proportions
Proving that EM improves the log probability of the training data
Using a Gaussian agreed distribution
What is the best variance to use?
Sending a value assuming a mixture of two equal Gaussians
Using another message to make random decisions
What is the best distribution?
EM as coordinate descent in Free Energy
The advantage of using F to understand EM
The indecisive means algorithm