Research Overview

I am a 3rd year PhD student in Computer Science at University of California, San Diego. My current interests are in analytics databases for large datasets, and machine learning. I am advised by Yoav Freund.

I lean toward the applied side of machine learning and statistics, so I'm focused on a few application domains at a time. The applications I'm currently interested in are smart grid data compression/prediction/anomaly detection, and the analysis of log data for online education.

Research Manuscripts

• Matthew Elkherj, Yoav Freund. A System for Sending the Right Hint at the Right Time. "To Appear in ACM: Learning at Scale 2014 (L@S-14). Atlanta, Georgia. March 2014. "
• Matthew Elkherj, Prasad Jayanti. Reader-Writer Exclusion Supporting Upgrade and Downgrade with Starvation Freedom. "Dartmouth Computer Science Technical Report TR2011-684, June 2011. (Undergraduate Thesis, High-Honors)"

Biography

I double-majored in Mathematics and Computer Science at Dartmouth. For the first few years I did research with Richard Granger in computational neuroscience and robotics, and build a robot simulation system used by a few AI research labs on campus. During my senior year, I worked on algorithms for distributed concurrent locks. This culminated in a thesis, referenced above. My source code on projects I've worked on, see my github profile. For information about industry experience, see my Linkedin page.