# What is deep learning?

Japanese: ディープラーニング

**In one line:** Deep learning is a machine learning method that uses multi-layer neural networks to learn the features of data step by step.

Deep learning is one way of doing machine learning. It stacks many layers of a “neural network.” A neural network copies how nerve cells in the brain connect. The layers are deep, so it is called “deep.” Take pictures. The first layer picks up lines and dots. The next finds shapes. Deeper layers find faces and objects. Step by step, it grasps more complex features. In the 2010s, it made image recognition much more accurate, and people took notice. Generative AI is built on top of it too. Learning deeply is the same for people and AI. It takes a lot of layers. And a lot of snacks.

**Common mix-up:** It is not separate from machine learning. Deep learning is one of the methods inside machine learning.

**FAQ**

- Is deep learning different from machine learning? — No. It is one of the methods within machine learning.

Topic: Tech
Published: 2026-11-15
Source URL: https://eli5.net/en/w/deep-learning
日本語: https://eli5.net/w/deep-learning
