Andrew Barto and Richard Sutton Win the Turing Award for Reinforcement Learning

ACM, the Association for Computing Machinery, has awarded the 2024 ACM A.M. Turing Award to Andrew G. Barto and Richard S. Sutton for their contributions to reinforcement learning. Their work laid the conceptual and algorithmic foundations of the field, influencing modern artificial intelligence.

Barto, Professor Emeritus at the University of Massachusetts Amherst, and Sutton, Professor at the University of Alberta and Research Scientist at Keen Technologies, have been recognised for research spanning decades. The Turing Award, often called the “Nobel Prize in Computing,” includes a $1 million prize funded by Google.

“Barto and Sutton’s work demonstrates the immense potential of applying a multidisciplinary approach to longstanding challenges in our field,” said ACM President Yannis Ioannidis. “Their contributions continue to shape AI and provide insight into how the brain works.”

Reinforcement learning (RL) focuses on training intelligent systems through reward-based mechanisms. The approach, inspired by psychology and neuroscience, builds on Markov decision processes, where agents learn optimal strategies through trial and error. In the 1980s, Barto and Sutton formalised RL as a general problem framework and introduced key algorithms, including temporal difference learning and policy-gradient methods.

Their 1998 textbook Reinforcement Learning: An Introduction remains a standard reference, cited over 75,000 times. Their ideas influenced the integration of RL with deep learning, leading to advancements such as AlphaGo’s victories over human Go players and reinforcement learning from human feedback (RLHF) used in ChatGPT.

“In a 1947 lecture, Alan Turing stated, ‘What we want is a machine that can learn from experience,’” said Jeff Dean, senior vice president at Google. “Reinforcement learning, as pioneered by Barto and Sutton, directly answers Turing’s challenge.”

RL has been applied in areas such as robotics, network congestion control, chip design, and global supply chain optimisation. Research also suggests RL models align with findings on dopamine system functions in neuroscience.

Barto and Sutton’s contributions continue to impact AI, with applications expanding across industries. Their recognition with the Turing Award highlights the lasting significance of reinforcement learning in computing and beyond.

This recognition follows a growing acknowledgment of AI’s role in advancing scientific discovery.

Last year, Demis Hassabis, CEO and co-founder of Google DeepMind, and John M. Jumper were awarded the Nobel Prize in Chemistry for their contributions to protein structure prediction through the AI system AlphaFold,  alongside David Baker, a professor at the University of Washington.

Geoffrey Hinton, known as the ‘Godfather of AI,’ was also awarded the Nobel Prize in Physics alongside John Hopfield for developing the Boltzmann machine, a neural network model inspired by statistical physics.