Neural Network

core

A computing system loosely inspired by the brain: layers of simple units (neurons) connected by adjustable weights. During training, the network tweaks millions or billions of these weights until its outputs match the examples it is shown. Every modern generative model — from image diffusion models to LLMs — is a neural network at its core. For creators, this matters because a network behaves the way it learned, not the way it was programmed: you steer it with prompts and settings, not code. Example: Imagen 3 is a neural network that learned to map text descriptions to images.

Related terms