Which statement about training data is true?

• A. It is used to train the model.
• B. It is used to validate performance on unseen data.
• C. It includes the correct labels for input data, guiding learning.
• D. It is identical to testing data.

Answer: (C)

In supervised learning, the training data provides input examples paired with their correct outputs (labels). Those labels tell the model what the true answer should be for each example, and the learning process uses them to compute errors and adjust the model's parameters. This labeling is what guides the model to learn the mapping from inputs to outputs, making the statement about including the correct labels the accurate description of training data.

While training data is indeed used to train the model, the key distinction is that the labels drive learning directly. Validation or test data, on the other hand, are used to evaluate performance on unseen data, not to guide learning. And training data is not the same as testing data.