Monthly Archives: April 2009

I had a discussion the other day about using the weights returned by boosting to do outlier detection. That is to say, the examples with high weight are hard to fit and are therefore likely to be mislabeled/outliers. So why … Continue reading →

… here’s a link to John Langford’s fantastic blog.

One of the things people learn during downturns is that it’s actually possible to look ab fab without spending a whole lot of money. Let’s try to apply the same enterprising spirit to posterior inference on mixture models. Today we’ll … Continue reading →