Reinforcement learning pioneers get 2024 Turing Award

The Association for Computing Machinery (ACM) has announced Andrew G Barto and Richard S Sutton as the recipients of the 2024 ACM A M Turing Award for their foundational contributions to reinforcement learning (RL). 

The Turing Award, often referred to as the "Nobel Prize of Computing," was founded in 1966 and comes with a $1 million prize sponsored by Google.

Barto is currently Professor Emeritus at the University of Massachusetts Amherst, while Sutton is a Professor of Computer Science at the University of Alberta and Research Scientist at Keen Technologies.

The two began their work in the 1980s, laying down the mathematical and algorithmic foundations for RL, one of the most critical approaches in artificial intelligence today.

Reinforcement learning is a process in which an AI agent learns to optimize its behaviour based on feedback in the form of rewards – a concept borrowed from the field of neuroscience and psychology, and worked on by legendary Alan Turing, the British mathematician and cryptographer. 

Barto and Sutton built their model upon the mathematical foundations of Markov decision processes (MDPs), which model decision-making in stochastic environments where outcomes are partly random and partly influenced by an agent's choices. 

One of the most influential aspects of their work is their 1998 textbook, 'Reinforcement Learning: An Introduction,' which remains the definitive reference in the field. Cited over 75,000 times, it has inspired thousands of researchers and shaped the AI landscape.

Although Barto and Sutton's algorithms were developed decades ago, RL gained widespread practical success in the last 15 years, particularly when combined with deep learning—a development spearheaded by 2018 Turing Award winners Yoshua Bengio, Geoffrey Hinton, and Yann LeCun. 

Reinforcement learning announced its arrival when AlphaGo, the AI system built on the technology, defeated human Go champions in 2016 and 2017. 

Since then, it has also been used in training robots to solve Rubik's cube and other complex tasks like optimising network congestion control, improving internet advertising strategies, and enhancing global supply chains. 

"Barto and Sutton’s work demonstrates the immense potential of applying a multidisciplinary approach to longstanding challenges in our field," ACM President Yannis Ioannidis said. 

"Research areas ranging from cognitive science and psychology to neuroscience inspired the development of reinforcement learning, which has laid the foundations for some of the most important advances in AI and has given us greater insight into how the brain works," he added.