Which of the following best describes reinforcement learning?

• A. Learns from labeled data to predict a target
• B. Learns patterns in unlabeled data
• C. Learns by interacting with an environment to maximize reward
• D. Uses predefined rules without data

Answer: (C)

Reinforcement learning centers on learning through interaction with an environment, where an agent takes actions, observes the results, and aims to maximize cumulative reward over time. The key idea is sequential decision making: each action changes the state and future rewards, so the agent continuously updates a policy that decides what to do in each situation. This distinguishes it from learning from labeled examples (supervised learning), discovering structure in unlabeled data (unsupervised learning), or following fixed, predefined rules without learning from feedback. In practice, you can think of a game-playing agent or a robot learning to navigate by trying moves, seeing which ones yield better long-term rewards, and gradually improving its strategy through trial and error.