This folder contains a file that is used on the MlClass from prof. Ng exemplifying the learning regularization using covariance matrix. Also to make it more clear I added a photo of a board showing the equation used as a learning algorithm. The only difference between this and the usual solution used on Ng's class is the dynamic regularization theorem used that sets the expected noise as a penality on the learning algorithm using covariance matrix.
Unfortunitly as the copyright isn't mine I can't share with others the full source code used to solve this problem, so one that wants to check the solution must have lesson 4 of Ng class. I strongly recommend to subscrible on his free Machine Learning class on coursera.org.