Open learning resources by Flow Math

25 visual lessons for reading linear algebra as a scene

Go beyond memorizing definitions and calculations. Follow each mathematical claim to the evidence it leaves in a moving 2D or 3D scene. Every lesson includes the original runnable notebook and a direct animation link.

8 units · 25 visual notebooks · free to read

Unit 1

Matrices & Gaussian elimination

Connect row and column pictures of a system to elimination, LU, and inverse matrices.

5 lessons
  1. Lesson 1.1

    Two views of a system

    The column view mixes columns c1 and c2 of A to make target b

    Read lesson
  2. Lesson 1.2

    Gaussian elimination

    Subtract twice the first row of A from the second to create a zero below the first pivot

    Read lesson
  3. Lesson 1.3

    Matrix columns

    Columns c1 and c2 of A are the destinations of basis vectors i and j

    Read lesson
  4. Lesson 1.4

    LU factorization

    L remembers elimination multipliers while U is the upper-triangular result

    Read lesson
  5. Lesson 1.5

    Inverse matrices

    When A moves space, x moves from (1, 1) to (3, 2)

    Read lesson
Unit 2

Vector spaces & transformations

Grow from linear combinations to independence, bases, the fundamental subspaces, and transformations.

7 lessons
  1. Lesson 2.1

    Vectors & scalars

    A vector is an arrow with both magnitude and direction

    Read lesson
  2. Lesson 2.2

    Vector addition

    Compare u and v from the same origin

    Read lesson
  3. Lesson 2.3

    Linear combinations

    A linear combination multiplies each vector by a coefficient and then adds them

    Read lesson
  4. Lesson 2.4

    Independence & dependence

    Neither a nor b is a scalar multiple of the other, so they provide two directions

    Read lesson
  5. Lesson 2.5

    Bases & coordinates

    Independent vectors such as b1 and b2 that generate every direction form a basis

    Read lesson
  6. Lesson 2.6

    Column space & null space

    A removes the z direction from three-dimensional space

    Read lesson
  7. Lesson 2.7

    Scaling & rotation

    Diagonal matrix S keeps the axis directions but doubles horizontal length

    Read lesson
Unit 3

Orthogonality, projection & least squares

Carry the right angle created by a dot product into projection and least-squares error.

3 lessons
  1. Lesson 3.1

    Dot products & orthogonality

    The dot product measures how much two vectors face the same direction

    Read lesson
  2. Lesson 3.2

    Projection

    A projection is the shadow of a vector on an axis or subspace

    Read lesson
  3. Lesson 3.3

    Least squares

    When target b is not on direction a, choose its closest shadow p

    Read lesson
Unit 4

Determinants

Read a determinant as an area or volume scale and watch a dimension disappear when it is zero.

2 lessons
  1. Lesson 4.1

    Determinants

    The determinant is the signed area made by two vectors

    Read lesson
  2. Lesson 4.2

    Area scale

    Before the transform, basis vectors i and j form a unit square of area 1

    Read lesson
Unit 5

Eigenvalues & eigenvectors

Start with directions preserved by a transform and see how diagonalization simplifies the action.

2 lessons
  1. Lesson 5.1

    Eigenvectors

    e keeps its direction through D, while q changes direction

    Read lesson
  2. Lesson 5.2

    Diagonalization

    The columns of P are eigenvectors whose directions are preserved by A

    Read lesson
Unit 6

Positive definite matrices & SVD

Read positive definiteness as energy and split SVD into rotate, scale, rotate.

2 lessons
  1. Lesson 6.1

    Positive definiteness

    A symmetric A is positive definite when xᵀAx is positive in every nonzero direction

    Read lesson
  2. Lesson 6.2

    Singular value decomposition

    SVD splits one matrix into a Vᵀ rotation, a Σ scaling, and a U rotation

    Read lesson
Unit 7

Matrix computation & stability

See how conditioning magnifies input error and how repeated computation converges toward one direction.

2 lessons
  1. Lesson 7.1

    Conditioning & sensitivity

    Two clearly separated lines have an intersection stable that stays stable under small changes

    Read lesson
  2. Lesson 7.2

    Iterative computation

    Each application of A halves the y component while preserving the x component

    Read lesson
Unit 8

Linear programming & game theory

Use constraint vertices and payoff matrices to explore optimization applications.

2 lessons
  1. Lesson 8.1

    Constraints & vertices

    Boundaries of linear inequalities become lines enclosing a feasible region

    Read lesson
  2. Lesson 8.2

    Game equilibrium

    G records the payoff for each pair of strategies

    Read lesson

01

About Flow Math

Flow Math is a visual mathematics project by Made by need that helps people understand equations as moving scenes. The public Linear Algebra edition connects matrices, vectors, bases, determinants, rank, null space, and systems to runnable notebooks and 2D or 3D scenes. It is not an official course of a school or educational institution.

02

How these lessons are made

Each lesson begins with one mathematical claim and the visual evidence needed to test it. It announces an operation, moves the real mathematical objects, and leaves the conclusion visible in the final scene. Playback, pause, rewind, and camera exploration let you inspect the result yourself.

03

Corrections and contact

Email us about mathematical errors, scene problems, accessibility, or privacy requests. For a reproducible issue, include your environment and a shared URL when possible.

Contact [email protected]