For people interested, this is an expanded version of the talk Stephen Boyd gave in KDD 2011
He shows how convex optimization is related to machine learning, specifically Support Vector Machines.
Sorry I marked it nsfw, it was a mistake.
[link] [5 comments]