Glen BersethView profile
Assistant Professor
Glen Berseth is an Assistant Professor in the Department of Computer Science and Operations Research (DIRO) at Université de Montréal and a core academic member of Mila – Quebec Artificial Intelligence Institute. He holds the prestigious Canada CIFAR AI Chair position and co-directs the Robotics and Embodied AI Lab (REAL). Previously, he was a postdoctoral researcher at Berkeley Artificial Intelligence Research (BAIR) working with Sergey Levine. Dr. Berseth's research focuses on solving sequential decision-making problems for real-world autonomous learning systems, with particular emphasis on human-robot collaboration, reinforcement learning, and various advanced learning paradigms including continual, meta, multi-agent, and hierarchical learning. His work bridges theoretical machine learning with practical robotics applications, aiming to create more capable and adaptable robotic systems. He has published extensively in top venues for robotics, machine learning, and computer animation. His recent publications demonstrate a strong focus on exploration strategies, representation learning, and efficient reinforcement learning algorithms. These works address critical challenges in scaling RL to real-world applications, improving generalization across different robot morphologies, and enabling autonomous learning without extensive human supervision. His research has practical implications for robotics, power grid control, and other complex systems requiring adaptive decision-making. Canada CIFAR AI Chair Dr. Berseth mentors a large group of graduate students across multiple institutions, supervising PhD and Master's candidates working on diverse aspects of robot learning and reinforcement learning. His lab receives significant research funding through his CIFAR AI Chair position and likely other grants supporting his robotics and AI research. The Robotics and Embodied AI Lab (REAL) he co-directs serves as a hub for interdisciplinary research at the intersection of machine learning and physical systems, fostering collaboration between computer scientists, roboticists, and domain experts.
