Bio and Current Research¶
I got my undergraduate in physics and mathematics from the University of Colorado in Boulder, CO; then decided to pursue graduate studies in Computer Science. I began studying CS with an emphasis on Machine Learning, and cluster validation methods in particular, at the University of British Columbia in September 2006. I was co-supervised by Kevin Murphy and Bertrand Clarke (Statistics Dept.). I am now beginning my Ph.D. in Statistics at the University of Washington.
During my master’s degree, my primary area of research was on unsupervised learning methods, particularly clustering stability analysis. Together with Bertrand Clarke, we developed a Bayesian approach to cluster stability analysis based on perturbing the clustering function. (See my Master’s thesis for details.). This allows fast and accurate stability assessments of model based clustering and may provide methods for automatic cluster verification and ranking. New and improved work based on this research is contained in On The Limits of Clustering in High Dimensions via Cost Functions. and A Bayesian Criterion for Cluster Stability..
Currently, I am in a PhD student in statistics at the University of Washington. I am working with Marina Meila on two different research directions. First, we are developing gravimetric inversion methods under a grant from the NGA. Essentially, we’re trying to detect underground features – caves, wells, tunnels, lava tubes, etc. – using precise measurements of the horizontal gravitational gradients on the surface. The current work involves detailed feature engineering, compressed sensing and sparse optimization methods. This has led to several of my code projects. Second, I am working on discrete optimization techniques for lattice and graph structures. The goal of this research is to develop efficient optimization methods for working with spatial and temporal statistical models.
Besides research and coding, my main hobby is bartending and mixology, and I bartend occasionally at Lucid Lounge.