معرفی
Richard S. Sutton is a pioneering researcher and Professor of Computing Science at the University of Alberta, where he serves as Chief Scientific Advisor for the Alberta Machine Intelligence Institute (Amii). He is also a Canada CIFAR AI Chair, Senior Fellow at CIFAR, and Research Scientist at Keen Technologies. Sutton is widely recognized as one of the founders of reinforcement learning, a field in which he continues to lead globally. His work has profoundly shaped modern AI research and applications.
Education:
- B.A., Psychology, Stanford University (1978)
- M.S., Computer Science, University of Massachusetts (1980)
- Ph.D., Computer Science, University of Massachusetts (1984)
Sutton's research focuses on identifying computational principles underlying intelligence and goal-directed behavior. He emphasizes learning from experience and extending reinforcement learning to create empirically grounded approaches to knowledge representation based on prediction. His work bridges artificial and natural intelligence, exploring how systems can predict and influence the world through learning, perception, action, and cognition. Sutton's research has produced foundational contributions including temporal-difference learning theory, actor-critic algorithms, the Dyna architecture, and Horde architecture.
His recent publications reveal a continued focus on reinforcement learning fundamentals while expanding into broader AI applications. The articles show strong emphasis on theoretical foundations, convergence analysis, and practical implementations of reinforcement learning algorithms. There's also evidence of applying these techniques to real-world problems like pandemic forecasting, suggesting growing interest in practical societal applications of his theoretical work.
Scientific Awards:
- 2024 Turing Award (with Andrew Barto)
- Fellow of the Royal Society (2021)
- Fellow of the Royal Society of Canada (2016)
- CAIAC Lifetime Achievement Award (2018)
- Outstanding Achievement in Research Award, UMass Amherst (2013)
Sutton has mentored approximately 60 early-career researchers, including notable figures like David Silver (lead researcher behind AlphaGo), Doina Precup (DeepMind Montreal), Adam White (Amii Fellow), and Cam Linke (CEO of Amii). His Reinforcement Learning & Artificial Intelligence Lab at the University of Alberta has become a global hub for RL research. Sutton is currently collaborating with John Carmack through Keen Technologies to accelerate AGI development, documented in part through 'The Alberta Plan' which outlines his vision for creating long-lived computational agents.
Sutton founded the Reinforcement Learning & Artificial Intelligence (RLAI) Lab at the University of Alberta and co-authored the seminal textbook 'Reinforcement Learning: An Introduction' with Andrew Barto. His work has been cited over 130,000 times and featured in major publications including Science, The Economist, New York Times, and Wall Street Journal. He is also known for his 'Tea Time Talks' tradition at Amii, fostering new ideas among researchers and students.
