Current doctoral candidate at Florida Atlantic University with a focus on Machine Learning and Fluid Mechanics. My background is in systems engineering studying underwater robotics and received my bachelors and masters in Ocean Systems Engineering. Since then, I've moved into Data Science and Physics for my dissertation where I found a passion for programming and computers.

Turbulent fluid dynamics are called the last great challenge of classical physics: the archetype of a chaotic nonlinear system where small changes can result in dramatically different outcomes. The difficulty posed by this problem coupled with the far reaching and valuable applications are why it is one of the biggest prizes in science and mathematics.. Simulations of this environment come with a heavy computational burden and generate large amounts of data. It is no wonder that fluid dynamics researchers ride the latest trends in data analysis including deep learning.

Traditional control schemes of turbulent flow bring together data analysis and and engineering controls for dimensionality reduction and feedback controls which rely on a priori knowledge of the system and fine-tuning. Deep Reinforcement learning (RL) is a branch of Deep Learning which gives the pattern finding agent the ability to take actions in a state in order to learn cause-and-effect relationships which is currently being applied to many topics from timing insulin application, to defeating the world champion at Go, to smart grid energy optimization. RL branches the two requirements of control schemes for turbulent fluids simulations, data analysis and feedback control, in a way that puts it in a unique position for success in turbulent flow control and why I chose to use it for my dissertation.