Michael Littman

Michael Littman discusses how interaction and teaching can dramatically improve the efficiency of RL and AI systems. He emphasizes the importance of agency — the ability of AI systems to actively seek out information and learn from interactions — and how this perspective can lead to more sample-efficient and practically useful learning algorithms.

Michael L. Littman is a University Professor of Computer Science at Brown University. He has made significant contributions to reinforcement learning, game theory, and AI education. He is known for his work on Markov games, partially observable Markov decision processes, and the integration of RL with natural language understanding.

He is a fellow of AAAI, ACM, and IEEE, and has received the AAAI Classic Paper Award. He served on the National Science Board and has been a strong advocate for AI literacy and education.