Speech Recognition

Unsupervised Deep Learning, in particular Generative Stochastic Networks and Generative Adversarial Networks.

Probabilistic Programming.

I worked at the Distributed Systems Group at Trinity College, Dublin the summer of 2012. I wrote C++ code to integrate CLIPS, a rules engine in C, and a local module of GRIDLAB-D in C++, so that reinforcement learning algorithms could be tested and simulated to work on smart power grid to increase their efficiency.

Designed and developed a sparse linear algebra package as a sub-module to Sympy which can handle matrices with symbolic elements. Used the C-extension GMPy for its number crunching for optimized performance.