What distinguishes supervised learning from unsupervised learning?

Supervised learning is characterized by the use of labeled data, which means that the training dataset includes input-output pairs where the output value (or label) is known. This allows the algorithm to learn a mapping from inputs to outputs, making it suitable for tasks such as classification and regression. The model is trained on this labeled data, enabling it to make predictions or classifications on new, unseen data based on the patterns it learned from the training set.

In contrast, unsupervised learning deals with data that does not have labeled outputs. This type of learning is used to find hidden patterns or intrinsic structures in the input data, such as clustering or association tasks. While unsupervised learning might require a significant amount of data to discover meaningful patterns, the defining characteristic that distinguishes it from supervised learning is not the amount of data but the absence of labels for the outputs.

The focus on labeled data in supervised learning is critical, as it directly impacts how the model is trained and evaluated. The recorded outcomes in the training process are essential for supervised learning algorithms to optimize their predictions over time.