🖥 Turing Award Goes to the Creators of Reinforcement Learning

The Turing Award, often called the "Nobel Prize of Computing," was awarded last week to Andrew Barto and Richard Sutton. They will share the $1 million prize. The scientists were pioneers of reinforcement learning—a method that underpins virtually all modern large language models.

ℹ️ Reinforcement learning is based on a "trial-and-error" approach: a system performs an action, receives positive or negative feedback, and gradually learns to select actions that yield greater "rewards."

In 1984, Sutton completed his PhD dissertation on reinforcement learning under Barto's supervision at the University of Massachusetts Amherst. They were considered eccentric researchers, attempting to teach machines to learn from their own experiences, drawing inspiration from studies in biology and psychology, including Edward Thorndike's experiments, which demonstrated that stimuli shape animal behavior.

In 2016, reinforcement learning—whose foundations were described by Barto and Sutton—gained worldwide recognition thanks to AlphaGo, a program that independently learned to play the board game Go better than any human.

📈 Today, reinforcement learning is used in advertising, energy optimization, finance, and chip design. It has become a key catalyst for the rapid progress of large language models. It is used to train reasoning models and create AI agents.

However, Sutton emphasizes that the methods used in today's large language models (LLMs) do not fully reflect the original concept he and Barto envisioned. For example, during training, systems like ChatGPT rely on goals set by humans. In contrast, their initial idea was for fully autonomous self-learning.

🏆 This is not the first time the Turing Award has been given for advancements in AI. In 2018, Yann LeCun, Yoshua Bengio, and Geoffrey Hinton received the award "for breakthroughs that have made deep neural networks a critical component of computing." Last year, Hinton shared the Nobel Prize in Physics with John Hopfield for their contributions to AI development.

Interviews with the Laureates:

👉 Geoffrey Hinton—on AI algorithms: "These 'creatures' are smarter than us!"

👉 Yann LeCun—on creating AI models with common sense

#news #machine_learning @hiaimediaen