Thursday, March 6, 2014

Abstract:
Everyone knows you ought to threat model, but in practical reality it turns out to be tricky. If past efforts to threat model haven't panned out, perhaps part of the problem is confusion over what works, and how the various approaches conflict or align. This talk is designed to help security pros, developers, and systems managers, all of whom will leave with both threat modeling lessons from Star Wars and a proven foundation, enabling them to threat model effectively ...
Read the full abstract and bio.

Tuesday, February 18, 2014

Abstract:
The Singular Value Decomposition (SVD) of matrices and the related Principal Components Analysis (PCA) are workhorses of machine learning and data analysis. They express a matrix in terms of eigenvectors, which are optimal in many ways; but they are linear combinations of all the input data, and thus they lack an intuitive physical interpretation, which is also of interest in many machine learning and data analysis applications ....
Read the full abstract