Which learning model requires input data paired with the correct output?

Supervised learning is characterized by the requirement of labeled input data, where each input is paired with the correct output. This form of learning is used to train models on a specific dataset to make predictions or classifications based on new, unseen data. In supervised learning tasks, the model learns to map input variables to the desired output by using the provided pairs of inputs and outputs, allowing it to understand the relationship between the two.

The success of supervised learning depends on the quality and quantity of the labeled data available for training. Once trained, the model can generalize from this data to predict outputs for new inputs. This contrasts with other learning models. Unsupervised learning does not use labeled data and instead looks for patterns or structures within the input data itself. Reinforcement learning involves agents that learn by interacting with an environment, using rewards and penalties rather than direct input-output pairs. Generative learning, meanwhile, refers to a method where the model learns to generate new data points based on the underlying distribution of the training data, also without direct input-output pairs.

The definitive characteristic of supervised learning is its reliance on explicit pairs of inputs and outputs, making it clear why it is the correct answer.