Some of my repos are currently set to private, but I'm happy to share: If you are interested in my work on reinforcement learning, machine learning techniques for solving inverse problems in science and engineering, or numerical approximation tools, please reach out to me at cameron DOT hickert AT gmail.
Two project samples are included here. The first is the PDF above detailing an exploration of convergence behavior of the proximal policy optimization (PPO) reinforcement learning algorithm. Another — a textual style transfer project utilizing OpenAI's GPT-2 language model with distributed model training via Horovod and MPI — can be viewed here. The information for the two projects' collaborators can be found in each.