AI Foundationspredict · compress · act

concepts → OpenQuestion

OpenQuestion

Continual learning

timeless1 connectiondraft

learning new things without overwriting the old — the thing fine-tuning still cannot do

Where it sits

The prism has six jobs across and eight layers down. Its primary cell is no mode × L4. Hatched cells cannot exist — a GPU does not learn, an institution does not infer.

Overlays: Safety & alignment.

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 pointContinual learningnothing depends on it yet — a leaf in the reading order

Where to read it

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

32Teaching Tasteact VIII · The Attention Revolution

Where it comes from

paperOvercoming Catastrophic Forgetting in Neural NetworksJames Kirkpatrick, Razvan Pascanu, Neil Rabinowitz, Joel Veness, Guillaume Desjardins, Alexander Rusu, Kieran Milan, John Quan, Tiago Ramalho, Agnieszka Grabska-Barwinska, Demis Hassabis, Claudia Clopath, Dharshan Kumaran, Raia Hadsell · 2017

Every connection

All 1 edges touching this node, grouped by relation family — the sections above are highlights from this list. Colours match the relation families inthe atlas.

Contrast · 1
trades off withFine-tuningAlgorithm

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.