Glossary · 1 · Foundations: how AI works
Neural network
Also known as: Deep learning, Artificial neural network
A neural network is a machine learning model made of layers of connected numeric units whose connection strengths — the weights — are adjusted during training so that the network maps inputs, such as words or pixels, to useful outputs. Networks with many layers are called deep learning.
- Beginner
- Intermediate
- Technical writers
- Technical marketers
- Technical project managers
- Developers
In one sentence
What a neural network is: layers of weighted connections, tuned during training, that underlie deep learning and every large language model.
Example
An image classifier passes the pixels of a photo through dozens of layers; early layers detect edges, later layers detect shapes such as a valve or a warning sign.
Why it matters on your learning path
- Technical writers: You don’t need the math. The key idea: the model’s knowledge is stored in weights, not in a database of documents it can look up.
- Technical marketers: “Deep learning” and “neural network” describe how a model is built; they say nothing about how good it is.
- Technical project managers: Neural networks need significant compute for training; most projects use pretrained models instead of training their own.
- Developers: Frameworks such as PyTorch and JAX, and libraries on Hugging Face, are the entry points for working with networks directly.