This function performs teh M-step of the EM algorithm. In the M-step we run chow-liu to get a new update of the tree and the big correlation matrix. The problem is that the tree does not always make sense so we need to replace it with a binary tree whose leaves correspond exactly to observed nodes. Luckily such a tree, giving exactly the same observed likelihood, always exists.
1 | M.step(S)
|
S |
covariance matrix over the whole vector (both observed and hidden variables) |
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.