I am a PhD student in Computer Science at UMass Amherst .
I'm particularly interested in machine learning as applied to learning causal models. I'm also interested in compiler design and implementation, predictability/unpredictability of computer systems, and quantum computing. In general, I like to play with data.
I'm comparing the performance of simple and more sophisticated classifier technologies when run on data drawn from a different distribution than the original training and test sets. This is motivated by the observation that the data on which classifiers are intended to run is usually temporally separated from the original training and test data, which increases the chances of changes in the distribution, such as population drift. This work is part of the Knowledge Discovery Laboratory with David Jensen .
I'm working with Eliot Moss in the Architecture and Language Implementation Laboratory to investigate the characteristics of performance of Java benchmarks with respect to the choice of garbage collector. Previous studies suggest that some aspects of computer performance may be chaotic, and garbage collection may result in chaotic memory references.
I graduated from Amherst College in 2011 with a B.A., Summa cum Laude with Distinction, in Computer Science . Professor Lyle McGeoch was my undergraduate thesis advisor.