Which approach focuses on learning from data by adjusting to labeled examples?

• A. Supervised learning
• B. Self-supervised learning
• C. Deep learning
• D. Mask learning

Answer: (A)

Supervised learning focuses on learning from labeled examples, where each input has a correct output provided. The model makes a prediction, a loss function compares that prediction to the true label, and the learning algorithm adjusts the model’s parameters to minimize that loss. With many labeled examples, the model learns to map inputs to the right outputs and can generalize to new data.

Self-supervised learning, in contrast, uses unlabeled data by creating its own learning signals from the data itself (such as predicting a missing part), so it doesn’t rely on human-provided labels to learn representations. Deep learning describes a broad family of methods that use deep neural networks and can be applied in supervised, self-supervised, or other learning setups. Mask learning isn’t a standard formal category for this concept; masking tasks are typically encountered within self-supervised or representation-learning contexts rather than a labeled-data–driven approach.