Interpretable Reinforcement Learning (FINS)
Poster presentation on interpretability methods for RL.
PI / Advisor: Dr. Damon Woodard
Institution / Department: Florida Institute for National Security (FINS), University of Florida
Timeframe: September 2024-April 2025
Research Focus
This project investigated interpretability techniques for reinforcement learning, focusing on how decision-making processes can be made more transparent without degrading performance.
Responsibilities
I developed a poster presentation that summarized methods for explaining RL policies, highlighting visualization strategies and evaluation metrics. The presentation was delivered as part of the UF Undergraduate Research Symposium and emphasized the importance of interpretability for real-world applications of RL.