Volunteer Computing is a means of harnessing idle cycles throughout the world toward solving important problems. The ALFA group (Anyscale Learning For All) currently has a system under development that can build a large-scale probabilistic model via volunteer computing. Volunteer computing requires strategies for efficient distribution of data; this project focuses on executing unsupervised learning by building a latent variable model that can identify clusters within the data. The aim is to expand upon Hidden Markov Models, and build other kinds of latent variable models. Ultimately the overall goal is to provide users – such as research scientists – with a broader range of choices than they currently have to analyze data.

I’ve worked at the MIT Media Lab implementing code-level tracking to better adapt mapping software to public needs, and at Intel, working on the shared virtual memory model for the Xeon Phi Coprocessor. Taking an undergraduate machine learning course really inspired my interest for large scale computing and machine learning. I’m excited to apply those skills to real-world applications such as volunteer computing.