Hello, I’m Samuel Neumann, a PhD student at the University of Alberta.
You can find my CV here.
Education and Research Interests
I earned my Bachelor of Science from MacEwan University in 2020, where I pursued a double major in Mathematics and Computing Science. My undergraduate research in machine learning was supported by two NSERC Undergraduate Student Research Awards (USRA). Upon graduation, I was awarded the Governor General’s Silver Medal for obtaining the highest GPA in my graduating class.
After completing my Bachelor of Science program, I enrolled in the Master of Science program at the University of Alberta under the supervision of Adam White. My graduate research focused on Reinforcement Learning (RL). My thesis introduced the Greedy Actor-Critic algorithm, an off-policy method designed to mitigate the hyperparameter sensitivity often found in popular algorithms like Soft Actor-Critic. My Master’s research was generously supported by an NSERC CGS-M, the Alberta Graduate Excellence Scholarship, and the Alberta Innovates Graduate Student Scholarship. I graduated from the Master of Science program in 2022.
I am currently a PhD student at the University of Alberta under the supervision of Adam White. My doctoral research focuses on the fundamental mechanics of Reinforcement Learning. I am particularly interested in hyperparameter dynamics and the ways in which these critical settings influence algorithmic behaviours such as stability, usability, and performance. My work aims to develop methods for automated tuning, specifically within the context of single-lifetime learning where agents must adapt to their environments in real-time. I see this as a key barrier to the widespread adoption of RL in many different sectors that could benefit from automation. By addressing this barrier, I hope to bridge the gap between RL in the lab and its practical, large-scale applications in complex industries.