AI Foundationspredict · compress · act

concepts → Architecture

Architecture

Generative adversarial network

also called GAN

timeless3 connections

a forger and a detective trained against each other — the recipe that made generated images real

Where it sits

The prism has six jobs across and eight layers down. Its primary cell is represent × L1. Hatched cells cannot exist — a GPU does not learn, an institution does not infer.

What must come first — and what it unlocks

Left to right is reading order, derived from the prerequisite_of edges. Nothing here is hand-ordered: the diagram is the graph.

nothing comes first — this is a starting pointGenerative adversarial networknothing depends on it yet — a leaf in the reading order

How it goes wrong — and what answers that

known failure modes
Mode collapsethe generator finds one output the critic cannot reject, and produces that forever

Where to read it

The chapter that introduces it, and any chapter that uses it again.

33Noise into Imagesact VIII · The Attention Revolution

Where it comes from

paperGenerative Adversarial NetworksIan Goodfellow, et al. · 2014

Every connection

All 3 edges touching this node, grouped by relation family — the sections above are highlights from this list. Colours match the relation families inthe atlas.

Lineage · 1
is attributed toIan GoodfellowPerson
Contrast · 1
trades off withDiffusion modelArchitecture
Failure · 1
fails byMode collapseFailureMode

This page is a projection of one node in src/data/concepts.ts. It has no prose file of its own — 521 declared edges produce all 340 of these pages. Edit an edge and both endpoints change.