Deep learning
딥러닝
A machine learning method that stacks neural networks many layers deep so they learn complex features of photos, speech and text.
Overview
Deep learning is a branch of machine learning. It uses neural networks with tens or hundreds of layers of computation. With photos, for example, early layers pick up simple features like lines and colors, and later layers pick up complex ones like eyes, noses and faces. It learns in steps.
It brought big gains in jobs that older methods could not handle well, such as recognizing photos, understanding speech, translating and writing. Today's chat AI and image generation AI are all built with deep learning. The cost is that it needs a great deal of data and computing hardware such as GPUs.
An everyday example
It is like a long assembly line in a factory. The first person picks the parts, the next fits them, and the next smooths them. The more hands it passes through, the finer the product.
Why it matters
Deep learning is why AI seems to have suddenly become smart. It is also why AI companies spend so much on chips and data centers, since deep learning uses so much computing.