I am a prospective graduate student. My research interests are in increasing neural network efficiency and capability by using a combination of biological inspiration and information theory.

I see inefficiencies in our current systems for processing information. These are being found faster and faster as our systems get larger and larger and weeding out inefficiencies becomes more incentivized.

These speed ups should lead to sustainable machine learning techniques that can assist us in a variety of ways without doing further damage to their environment and being predictable in their unpredictability.

My current work on the information theory side is capturing abstractions at different levels for use in tasks needing human abstraction and objectification using cnn’s and mlp’s.

On the biological front, I’m trying to move to a system of processes that can be created to advance machines using evolution. Currently, these systems exist away from the state of the art in machine learning. If evolutionary mechanisms can assist in online-reinforcement-learning, they could be employed by and compared to state-of-the-art systems.