COWS and Their Hybrids: Customized Orthogonal Weights
Abstract: Particle physicists developed an algorithm called COWs (Customized Orthogonal Weights) for separating signals from backgrounds in certain experiments. We look at COWs from a statistical perspective. Then we consider several extensions of the method. In particular, a modified version of the method leads to a robust method for estimating arbitrary mixtures of conditionally independent distributions.
This is joint work with Chad Schafer and Mikael Kuusela.
Speaker Bio: Larry Wasserman is University UPMC Professor of Statistics and Data Science at Carnegie Mellon University. He is also Professor in the Machine Learning Department. He is a member of STAMPS (STAtistical Methods for the Physical Sciences).
Zoom link: https://utexas.zoom.us/j/84254847215