معرفی
Lucas Lehnert is an Assistant Professor in the Department of Computer Science at the University of Saskatchewan, specializing in Artificial Intelligence and Reinforcement Learning (RL). His research focuses on how intelligent systems can learn to solve complex decision-making tasks through representation learning, abstraction mechanisms, and lifelong learning strategies. He also explores applications of AI/RL in scientific and engineering domains.
Education: PhD in Computer Science (Brown University, 2021), MSc (McGill University, 2016), BSc (McGill University, 2014). Postdoctoral positions included Meta's FAIR team (2022–2024) and the Mila Quebec AI Institute (2021–2022).
Research interests include reinforcement learning fundamentals, generative AI reasoning, exploration strategies, and reward-predictive representations. His work bridges model-based and model-free RL paradigms, emphasizing scalable and generalizable solutions.
Awards include the Best Student Workshop Paper Award (2017) and an NIMH training grant in cognitive neuroscience. His research has been published in top conferences like NeurIPS, ICML, and ICLR.
He advises graduate students in RL and collaborates on projects involving transformer-based planning, exploration algorithms, and multi-agent systems. Current work includes developing SearchFormer for efficient planning tasks and exploring maximum entropy exploration methods.
