Michael is a first-year doctoral student in the Department of Computer Science. He is interested in the field of deep reinforcement learning and its applications to human-robot interactions, as well as the visualization of neural networks. Drawing from his background in neuroscience, molecular biology and mathematics, his goal is to develop novel methods for addressing the challenges in creating seamless human-robot interactions.
Michael's current research focus is on developing a visualization tool that will enable users to readily assess differences amongst neural network algorithms.
Publications
Iuzzolino, Michael L., Michael E. Walker, and Daniel Szafir. "Virtual-to-Real-World Transfer Learning for Robots on Wilderness Trails." In Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2018). IEEE, 2018. (Madrid, Spain – Oct. 1-5, 2018).