AI Glossary

Reinforcement learning

Also called RL

What does “Reinforcement learning” mean?

Reinforcement learning: Teaching a system through trial and error: it takes actions, receives rewards or penalties, and gradually learns which choices lead to better outcomes. It is used in game-playing and robotics, and to refine how chat models respond after their initial training.

Reinforcement learning visual guide: Teaching a system through trial and error: it takes actions, receives rewards or penalties, and gradually learns which choices lead to better outcomes. It is used in game-playing and robotics, and to refine how chat models respond after their initial training.
Visual guide Open full size (new tab)

What is an example of “Reinforcement learning”?

DeepMind's AlphaGo used reinforcement learning, playing many games against itself to improve at Go.

Keep learning

See also

Supervised learningTermReinforcement learning from human feedbackTermReward modelTermTrainingTerm

Same topic

Reinforcement learning from verifiable rewardsTermSelf-supervised learningTermPromptTermProbabilityTermPredictionTerm