A book by M. M. Akbar
Seeing Linear Algebra
See the math behind the model.
A geometry-first guide to the linear algebra behind machine learning, for people who can already train a model but want to see the math instead of just trusting it.
The full first chapter, figures and all, straight to your inbox. No spam, just the occasional update, and you can leave anytime.
Chapter 1 is heading to your inbox now. If it does not arrive in a few minutes, check your spam folder.
The book is being written in the open. It lands on Leanpub soon, where you'll be able to read it as it grows and name your price.
rotate · stretch · rotate
Seen, not memorized
Every idea arrives as a picture first and the formula second, so the math stops feeling like incantation.
Tied to real ML
Each concept earns its place by unlocking something you already do, from embeddings to attention.
Honest about the limits
Classical mathematics, told straight, with the caveats most courses quietly skip.
Inside the first chapter
- Why linear algebra is really the geometry of data, not just grids of numbers.
- Why a dataset is a cloud of points, and why its shape is the whole game.
- Where it all leads: every matrix as rotate, stretch, rotate, the SVD.
Five chapters are already drafted and fully illustrated. The first one is yours, free.