Reinforcement learning is a type of machine learning where an agent learns by interacting with an environment. It receives rewards for correct actions and penalties for wrong ones, gradually improving its behavior through trial and error.

‍