Which subset of machine learning involves neural networks with many layers?

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

Answer: (A)

Deep learning uses neural networks with many layers to automatically learn hierarchical representations from data. Each layer transforms the input into a more abstract and complex representation, so the network can move from raw data like pixels or tokens to meaningful concepts such as objects in an image or the meaning of a sentence. This depth enables the model to capture intricate patterns that shallower models often miss, which is why it shines in tasks like computer vision, speech, and natural language processing. While supervised learning involves labeled data and self-supervised learning relies on creating learning signals from data itself, deep learning specifically refers to the architecture of many-layer neural networks. A model can be trained in various ways, but the defining feature is the multi-layer structure that learns rich, hierarchical features.