Room 6 · The Awards · 2025

Barto and Sutton, Turing Award 2024

Barto and Sutton, Turing Award 2024 (Andrew Barto and Richard Sutton, 2025): Learning by trial, error and reward.

Andrew Barto and Richard Sutton received the 2024 Turing Award, announced in 2025, for developing the conceptual and algorithmic foundations of reinforcement learning: agents that learn by trial, error and reward. Those ideas are behind systems like AlphaGo and techniques like the human-preference training of Room 5.

In March 2025 the ACM announced that the 2024 Turing Award would go to Andrew Barto and Richard Sutton "for developing the conceptual and algorithmic foundations of reinforcement learning."

Reinforcement learning is learning from consequences. An agent, which can be a program or a robot, acts in an environment, receives rewards or penalties, and gradually learns which actions lead to the most reward over time. The idea has roots in animal psychology and in early machines like the SNARC of Room 2, but it lacked a solid mathematical and computational framework.

Barto, at the University of Massachusetts Amherst, and Sutton, who began as his doctoral student, built that framework from the late 1970s onward. They connected learning by trial and error with tools from optimal control, the mathematics of making a sequence of decisions. One of their central ideas is temporal difference learning, which Sutton formalized in 1988: an agent improves its predictions by comparing what it expected with what it sees one step later, without waiting for the final result. Their textbook, Reinforcement Learning: An Introduction, first published in 1998 and expanded in 2018, trained generations of researchers.

The applications came later. In 1992 a backgammon program using temporal difference learning reached the level of top players. DeepMind's Atari player and AlphaGo in Room 4 combined these ideas with neural networks, and reinforcement learning from human feedback, in Room 5, is how chat models learn to follow instructions.

In 2019 Sutton wrote a short essay, "The Bitter Lesson," arguing that over the history of AI, general methods that scale with computation have repeatedly beaten approaches built on human knowledge. It became one of the most discussed texts in the field.

Ask your AI: summarize "The Bitter Lesson" in three sentences, then give the strongest argument against it.

Did you know? Sutton wrote the essay “The Bitter Lesson”, arguing that general methods that scale with compute end up outperforming those built on human-written knowledge.

Source: awards.acm.org/about/2024-turing

See it in the museum →

← Nobel Prize in Chemistry 2024

All of Room 6: The Awards

Cite this: Nazareno, I. (2026). Barto and Sutton, Turing Award 2024 (2025). New Mind Museum. Retrieved November 17, 2026, from https://newmindmuseum.com/museum/barto-and-sutton-turing-award-2024-2025/