COURSE 04L2100% FREE
Verified 2026-08-10

Scalars, vectors, and matrices

AI math starts with three picture-friendly ideas: - **Scalar:** one number (temperature = 72) - **Vector:** an ordered list of numbers (a point or arrow) - **Matrix:** a table of numbers (rows × columns)

What it is

AI math starts with three picture-friendly ideas:

  • Scalar: one number (temperature = 72)
  • Vector: an ordered list of numbers (a point or arrow)
  • Matrix: a table of numbers (rows × columns)

Why it matters

Almost every model stores weights as vectors and matrices. You do not need to love proofs—but you do need to recognize the shapes so code and diagrams stop looking like alien glyphs.

How it works (plain)

A vector can mean a position, a direction, or a list of features (height, weight, age). A matrix can mean a grid of pixels, a spreadsheet of features, or a transformation that rotates/stretches vectors.

Everyday example

A shopping list with quantities is a vector of counts. A weekly grid of hours worked is a matrix.

Try it

Write a 3-number vector that describes a fruit (weight, sweetness 1–10, price). That is a tiny feature vector.

Myths

⚠️ Myth: You must finish a full linear-algebra course before learning AI.
✓ Reality: Picture-first intuition unblocks most Wave 1 chapters; deeper math is optional depth.
⚠️ Myth: Matrices are only for geniuses.
✓ Reality: They are organized tables with rules for combining them.

Sources