About
Zachary Roch is a second year Ph.D. student studying Computer Engineering with a specialization in Intelligent Systems and Machine Learning at the University of Central Florida (UCF). He is one of the first students brought on under the Institute of Artificial Intelligence (IAI), a cross-disciplinary initiative between multiple colleges with the goal of fostering collaboration to advance various artificial intelligence (AI) disciplines. Under this initiative, Zach’s research interests surround the theoretical foundations and analysis of both single-agent and multi-agent robust reinforcement learning under the average-reward criterion (long-term horizon), taking his experience in the finance industry to take a game theoretic approach using the Markov game formulation. Currently, he is working on reinforcement learning with human feedback (RLHF) and cross-domain adaptation.
Calling Zachary’s academic journey to reach his current work unconventional would be an understatement. Originally obtaining a Bachelors in Science in Business Administration with a concentration in finance at UCF in 2020, Zachary took his skills to work as a stockbroker at Fidelity Investments where he obtained his Series 7 and Series 63 licenses. While here, he met with senior leadership and voiced his interest in pivoting towards the fund management side, and it was this meeting which led him to an epiphany that he did not possess the necessary technical skills to accomplish this goal. Given this, he took a leap of faith to return to UCF as part of the introductory cohort to pursue a Master of Science in Financial Technology (FinTech), the first program in the state of Florida to be offered in collaboration between the College of Business and the College of Computer Science. Through this program, Zachary gained proficiency working with Natural Language Processing (NLP), distributed ledger technology, and machine learning through development of various FinTech applications and models. During this program, Zachary took a special interest in data structures, algorithms, and computational complexity. This interest ultimately led him to his current program and research, where he currently has one paper accepted at the International Conference on Machine Learning (2025) and few more papers under review.
When Zach is not in the lab, he enjoys reading, visiting with family and friends, travelling, and gaming. He also spends plenty of time with his cat, Pepper, who listens to him talk through his code with the kind of silent skepticism that usually reveals a bug. Feel free to reach out to discuss reinforcement learning, blockchain technology, or to ask for Pepper to help review your code.
Papers
See details →
A Reduction Framework for Distributionally Robust Reinforcement Learning under Average Reward
Zachary Roch, George Atia, and Yue Wang
See details →
Provably Sample-Efficient Robust Reinforcement Learning with Average Reward
Zachary Roch, Chi Zhang, George Atia, and Yue Wang
See details →
Distributionally Robust Markov Games with Average Reward
Zachary Roch and Yue Wang
See details →
Blockchain Services for Digital Government: An Exploration of NFT Applications in the Metaverse
Zachary Roch and Ramya Akula
I am always open to discussing new ideas and collaborations in Robust Reinforcement Learning and Multi-agent Systems. Feel free to reach out if you have any questions about my work. Thank you!