// ✦ Diffusion, Flow Matching & Generative Models
VQ-VAE (Neural Discrete Representation)
Encodes images or audio into discrete codebook tokens.
VQ-VAE 离散表示 2017 Intermediate core
IN PLAIN WORDS
Encoder outputs snap to the nearest codebook vector, turning signals into discrete tokens that Transformers can model like text. Basis of image tokenizers and audio codecs.
Used in: Image/audio tokenizers, DALL·E 1
TIMELINE · prerequisites → this → what builds on it
01Original paper
Neural Discrete Representation Learning
Aaron van den Oord, Oriol Vinyals, Koray Kavukcuoglu · 2017
arXiv ↗ ✓ Verified · arXiv API
02Where it sits
Learn the left column first; the right column builds on this one.
Learn first
Related:Transformer & Self-Attention
Left: prerequisites · right: builds on this · dashed: related. Full map →
03Learning resources
Official docs, explainers, courses and reference code. All links checked.
Explainers
04Where the course teaches it
For AI / data
- Verification: verified, checked 2026-10-21.
- JSON: /api/learn/algorithms.json (id:
vq-vae)