A generative model is like a toy that can copy things it sees and make new ones from them.
Imagine you have a box full of different colored blocks. You show the box a red block, and then it makes another one just like it. But if you show it many red blocks, or even a pattern of blocks, it can start making its own patterns, maybe a row of red and blue blocks, or even a tower that looks like a building.
How It Works
Making New Things
Sometimes, generative models are used to make drawings, music, or even stories. They look at lots of examples first, then start creating their own, just like how you learn to draw from copying pictures in a book, and then you can make up your own shapes and scenes.
Examples
- A generative model is like a magic box that creates new drawings based on the ones you show it.
- You teach it how to paint, and it starts creating brand-new pictures.
Ask a question
See also
- How do large language models learn to talk like humans?
- How do large language models like ChatGPT actually learn?
- What do AI models balance between long-term predictions?
- Why do AI models sometimes 'hallucinate' or invent facts?
- What is Natural language processing (NLP)?