Andrew Barto and Rich Sutton, pioneers of reinforcement learning, have been honored with the Turing Award for their groundbreaking work in machine learning from experience. Initially deemed unfashionable, their approach has gained traction and is now integral to AI applications, including chatbots like ChatGPT and Google's AlphaGo. Through reinforcement learning, machines are able to learn tasks by receiving feedback, demonstrating applications in diverse fields such as robotics, finance, and language modeling. Sutton emphasizes the importance of human-directed goals in shaping modern AI development.
When this work started for me, it was extremely unfashionable. It's been remarkable that [it has] achieved some influence and some attention. Barto recalls with a smile.
Reinforcement learning was perhaps most famously used by Google DeepMind in 2016 to build AlphaGo, a program that learned for itself how to play the incredibly complex and subtle board game of Go to an expert level.
The same method is also being used to train AI models to mimic human reasoning, and to build more capable AI agents. Sutton acknowledges this progression.
Sutton notes, however, that the methods used to guide LLMs involve humans providing goals rather than an algorithm learning purely through its own exploration.
Collection
[
|
...
]