Which type of learning is characterized by labeled training data?

Supervised learning is characterized by the use of labeled training data, where each training example is paired with an output label. This approach allows the model to learn the relationship between inputs and their corresponding outputs. The labeled data provides a ground truth that the learning algorithm can use to correct its predictions, thereby guiding the model toward better performance on unseen data.

In supervised learning, the algorithm's goal is to learn a mapping from inputs to outputs based on the training data it receives. The process involves using these labeled examples to minimize the difference between predicted outputs and the actual labels, often through a method called loss minimization. This type of learning is widely used in applications such as classification and regression tasks, where knowing the correct outcome is essential for evaluating model accuracy.

In contrast, the other types of learning features different methodologies and data handling: Unsupervised learning does not use labeled data and focuses on identifying patterns or structures within unlabeled datasets. Reinforcement learning operates on the principle of learning from interactions with an environment to maximize cumulative rewards, without requiring labeled examples. Deep learning, while it can be applied within the context of supervised learning and utilizes neural networks, is more of a technique or subset within machine learning rather than a distinct learning type defined by labeled data.