When most people first encounter matrices, they see them as rectangular grids of numbers.

For example:

$$
A=
\begin{bmatrix}
2 & 0 \\
0 & 3
\end{bmatrix}
$$

At first glance, this appears to be nothing more than a convenient way of organizing numbers.

However, one of the most important insights in linear algebra is that a matrix is much more than a table of numbers.

A matrix is actually a function.

Understanding this idea completely changes how we think about linear algebra, quantum mechanics, machine learning, computer graphics, and modern mathematics.

What Is a Function?

A function takes an input and produces an output.

For example:

$$
f(x)=2x
$$

If the input is:

$$
x=3
$$

then the output is:

$$
f(3)=6
$$

A function is simply a rule that transforms one object into another.

A Matrix Does Exactly the Same Thing

Consider the matrix:

$$
A=
\begin{bmatrix}
2 & 0 \\
0 & 3
\end{bmatrix}
$$

and the vector:

$$
x=
\begin{bmatrix}
1 \\
2
\end{bmatrix}
$$

Multiplying the matrix by the vector gives:

$$
Ax=
\begin{bmatrix}
2 \\
6
\end{bmatrix}
$$

Notice what happened.

The matrix accepted an input vector and produced an output vector.

This is exactly what a function does.

The matrix acts like a machine.

Input:

$$
\begin{bmatrix}
1 \\
2
\end{bmatrix}
$$

Output:

$$
\begin{bmatrix}
2 \\
6
\end{bmatrix}
$$

Therefore the matrix is behaving as a function.

The Formal Perspective

In linear algebra, a matrix defines a function

$$
T:\mathbb{R}^n \rightarrow \mathbb{R}^m
$$

through the rule

$$
T(x)=Ax
$$

where:

• $$x$$ is the input vector
• $$A$$ is the matrix
• $$Ax$$ is the output vector

Thus every matrix defines a mapping from one vector space to another.

Not Every Function Is a Matrix

Consider the function:

$$
f(x)=x^2
$$

This is a perfectly valid function.

However:

$$
f(2x)=4x^2
$$

while

$$
2f(x)=2x^2
$$

These are not equal.

Therefore the function is not linear.

Matrices represent only a special class of functions known as linear functions.

What Does Linear Mean?

A function is linear if it satisfies two conditions.

Additivity

$$
T(u+v)=T(u)+T(v)
$$

Homogeneity

$$
T(cu)=cT(u)
$$

for every scalar $$c$$.

Every matrix transformation satisfies these properties.

That is why matrices and linear functions are essentially the same thing.

A Matrix as a Formula

Consider the function

$$
f(x,y)=
\begin{bmatrix}
2x+y\\
x+3y
\end{bmatrix}
$$

This can be written as:

$$
\begin{bmatrix}
2 & 1\\
1 & 3
\end{bmatrix}
\begin{bmatrix}
x\\
y
\end{bmatrix}
$$

The matrix is simply a compact way of expressing the function.

The matrix and the function contain exactly the same information.

The Geometric Interpretation

The most beautiful way to think about matrices is geometrically.

A matrix transforms space.

It can:

• Stretch space
• Compress space
• Rotate space
• Reflect space
• Shear space

The vectors are points in space.

The matrix tells us where those points move.

For example:

$$
A=
\begin{bmatrix}
2 & 0\\
0 & 1
\end{bmatrix}
$$

doubles every x-coordinate while leaving every y-coordinate unchanged.

A circle becomes an ellipse.

The matrix is acting as a geometric transformation of the entire plane.

Where Matrices Really Come From

Interestingly, modern mathematics often begins with functions rather than matrices.

Mathematicians first define a linear transformation:

$$
T:V \rightarrow W
$$

between vector spaces.

Only after choosing coordinates do they represent that transformation using a matrix.

The process is:

Linear Transformation

↓

Choose a Basis

↓

Matrix Representation

This means that the matrix is not the fundamental object.

The underlying function is.

The matrix is merely a convenient representation of that function.

Why This Matters

This perspective appears throughout modern science.

Machine Learning

Neural network layers are matrices acting on vectors.

Each layer is a function transforming data into a new representation.

Computer Graphics

Rotations, scaling, and 3D rendering are matrix transformations.

Matrices move objects through virtual space.

Quantum Mechanics

Quantum states are vectors.

Quantum operators are matrices acting on those vectors.

The matrix represents a physical transformation of the quantum state.

Functional Analysis

Linear operators generalize matrices to infinite-dimensional spaces.

The same idea survives, even when the matrix itself becomes infinitely large.

The Deepest Insight

Most students begin by thinking:

A matrix is a table of numbers.

Eventually they learn:

A matrix is a linear transformation.

But the deepest understanding is:

A matrix is a function that has been written in coordinates.

The numbers themselves are not the most important part.

The true object is the transformation—the rule that takes one vector and turns it into another.

Once this idea clicks, linear algebra stops being a collection of matrix manipulations and becomes the study of functions acting on spaces.

That shift in perspective is one of the most important conceptual breakthroughs in all of mathematics.