Generative Models — Taxonomy & History
Every image model, text model, video model, and 3D model fits in one of five buckets. Pick the wrong bucket and you will fight the math for weeks. Pick the right one and the field's last twelve years of progress stacks cleanly in your head. A generative model does one job: given training samples drawn from some unknown distribution pdata(x), output new samples that look like they came from the same distribution. Faces, sentences, MIDI files, protein structures — all the same problem if you squint. The rub is that pdata lives in a space with millions of dimensions (a 512x512 RGB image is 786k dimensions), the samples sit on a thin manifold inside that space, and you only have maybe 10M examples. Brute-forcing the density is hopeless. Every generative model is a compromise that trades one hard problem for a slightly less hard one. Five families have survived the last twelve years. Knowing which compromise each family makes tells you why it wins on some tasks and collapses on others. Five families of generative models — taxonomy by what they model 1. Explicit density, tractable. Write log p(x) as a sum you can actually evaluate. Autoregressive models (PixelCNN, WaveNet, GPT) factorize p(x) = ∏ p(xi | x<i). Normalizing flows (RealNVP, Glow) build p(x) as an invertible transform of a…
Generative Models — Taxonomy & History: Every image model, text model, video model, and 3D model fits in one of five buckets. Pick the wrong bucket and you…
This free lesson is part of the AI Engineering from Scratch curriculum. Read the full explanation, run the lesson code, and verify the result in the interactive reader or from the repository source.
Browse the complete course catalog or open this lesson on GitHub.