Projects

Improving PILCO with a Gaussian Process

PILCO is Model-based reinforcement learning algorithm used to solve control problems from scratch by utilizing Gaussian Process (GP) inference.

During my senior year of High School, I investigated replacing the GP with a Neural Network GP which is theoretically equivalent in an infinite limit for my AP Research Project.

You can view the paper Here (opens in a new tab) and the code for the project Here (opens in a new tab).