Which type of learning technique is 'reinforcement' ?

Reinforcement Learning (RL) is a subfield of machine learning where an autonomous agent learns to make decisions by interacting with an environment. It is fundamentally a feedback-based learning technique. In this process, the agent takes actions and receives feedback in the form of rewards or penalties. The goal of the agent is to maximize the cumulative reward over time through trial and error.

• Supervised Learning (e.g., classification) relies on labeled datasets.
• Unsupervised Learning identifies patterns in unlabeled data.
• Reinforcement Learning differs because it does not require pre-labeled data; instead, it relies on the dynamic feedback loop between the agent's actions and the environment's response.

Common applications include robotics, game playing (like AlphaGo), and autonomous driving, where the system must learn optimal behavior through continuous feedback.