ALGEBRALinear AlgebraMathematics Calculator
ker

Null Space (Kernel)

Null(A) = {x : Ax = 0}. The set of all solutions to the homogeneous system. dim(Null(A)) = nullity = n − rank(A). Basis from free variables in RREF.

Concept Fundamentals
Ax = 0
Definition
n − rank
Nullity
rank + nullity = n
Rank-nullity
from free vars
Basis
Compute Null SpaceKernel of A; basis vectors

Why This Mathematical Concept Matters

Why: Null space describes redundancy in Ax=b. Zero eigenvalue eigenvectors lie in null space. Essential for understanding linear maps.

How: Reduce A to RREF. Identify free columns. For each free variable, set it to 1 (others 0) and back-substitute to get basis vector.

  • ●Null(A) = {0} ⟺ A full column rank.
  • ●rank(A) + nullity(A) = n.
  • ●Basis size = nullity.
Sources:Gilbert Strang, Linear AlgebraWolfram MathWorld

Null Space Calculator

×matrix
⟪
⟫
⟨
⟩

What is the Null Space?

The null space (or kernel) of a matrix A is the set of all vectors x such that Ax = 0. In other words, it's the solution set to the homogeneous linear system Ax = 0.

Geometrically, the null space represents the set of vectors that are mapped to the zero vector by the linear transformation represented by the matrix A.

Formally, for an m×n matrix A, the null space is defined as:

Null(A)={x⃗∈Rn:Ax⃗=0⃗}\text{Null}(A) = \{\vec{x} \in \mathbb{R}^n : A\vec{x} = \vec{0}\}

Key Properties of the Null Space

Subspace Property

The null space is a subspace of Rn, where n is the number of columns in the matrix A. This means it contains the zero vector and is closed under addition and scalar multiplication.

Dimension

The dimension of the null space (nullity) equals n - rank(A), where n is the number of columns in A. This dimension represents the number of linearly independent vectors in any basis for the null space.

Rank-Nullity Theorem

The sum of the dimension of the null space and the rank of the matrix equals the number of columns in the matrix:

textrank(A)+textnullity(A)=n\\text{rank}(A) + \\text{nullity}(A) = n

This fundamental theorem establishes a direct relationship between a matrix's ability to map vectors to zero and its ability to span the output space.

Relation to Invertibility

A square matrix A is invertible if and only if its null space contains only the zero vector. For an invertible matrix, the system Ax = 0 has only the trivial solution x = 0.

Null Space of Products

For matrices A and B where the product AB is defined:

• If x is in Null(B), then Bx = 0, so ABx = A(0) = 0, which means x is in Null(AB)

• Therefore: Null(B) ⊆ Null(AB)

Null Space and Column Space

The null space of A is orthogonal to the row space of A. This means every vector in Null(A) is perpendicular to every vector in Row(A).

Geometric Interpretation

The null space has several important geometric interpretations that help visualize its significance:

Kernel of a Linear Transformation

If we think of a matrix A as a linear transformation TA: Rn → Rm, then the null space represents all vectors that get mapped to the origin (zero vector) by this transformation.

Example: In a 2D projection onto the x-axis, all points on any line parallel to the y-axis map to the same point on the x-axis. The null space consists of all vectors pointing in the y direction, as these get "squashed" to zero.

Solution Space to Homogeneous System

The null space represents all solutions to the homogeneous system Ax = 0. Geometrically, this is a subspace passing through the origin in Rn.

• For a 3×3 matrix with nullity 1: The null space is a line through the origin in R3

• For a 3×3 matrix with nullity 2: The null space is a plane through the origin in R3

Information Loss and Degeneracy

The null space represents the "information loss" in a transformation. A non-trivial null space means multiple input vectors map to the same output vector, making the transformation non-invertible.

Example: In image processing, if a transformation has a non-trivial null space, it means some features of the original image are completely lost in the transformed version, making perfect reconstruction impossible.

How to Find the Null Space

Step 1: Convert to RREF

Transform the matrix to its Reduced Row Echelon Form (RREF) using Gaussian elimination. This gives a standardized form where all pivot positions have leading 1's, and all columns with leading 1's have zeros elsewhere.

Step 2: Identify Pivot and Free Variables

Pivot columns in the RREF correspond to basic variables, while non-pivot columns correspond to free variables. The number of free variables equals the dimension of the null space.

Step 3: Express Basic Variables in Terms of Free Variables

From the RREF, express each basic variable in terms of the free variables by solving the system Ax = 0. This gives the parametric form of the null space.

Step 4: Create Basis Vectors

For each free variable, create a basis vector by setting that free variable to 1 and all other free variables to 0, then compute the corresponding values for the basic variables. The collection of these vectors forms a basis for the null space.

Example Process

Consider the matrix:

(13−226−4)\begin{pmatrix} 1 & 3 & -2 \\ 2 & 6 & -4 \end{pmatrix}

Step 1: Convert to RREF

(13−2000)\begin{pmatrix} 1 & 3 & -2 \\ 0 & 0 & 0 \end{pmatrix}

Step 2: Identify pivot (column 1) and free variables (columns 2 and 3)

Step 3: Express the basic variable x₁ in terms of the free variables x₂ and x₃

x1+3x2−2x3=0  ⟹  x1=−3x2+2x3x_1 + 3x_2 - 2x_3 = 0 \implies x_1 = -3x_2 + 2x_3

Step 4: Create basis vectors by setting each free variable to 1 with others at 0

For x2=1,x3=0:(−310)\text{For }x_2=1, x_3=0: \begin{pmatrix} -3 \\ 1 \\ 0 \end{pmatrix}
For x2=0,x3=1:(201)\text{For }x_2=0, x_3=1: \begin{pmatrix} 2 \\ 0 \\ 1 \end{pmatrix}

These two vectors form a basis for the null space, which is a 2-dimensional plane in R³.

Real-World Applications of the Null Space

Linear Systems and Engineering

The null space provides the homogeneous solutions to a linear system Ax = b. In engineering, this represents the degrees of freedom in a system.

Example: In structural analysis, the null space of the equilibrium matrix contains all self-stress states, where internal forces can exist without external loads.

Image Processing and Computer Vision

In image processing, the null space contains all transformations that result in no visible change to an image, which is critical for image compression algorithms.

Example: In JPEG compression, the null space of the discrete cosine transform matrix helps identify frequency components that can be discarded without significant visual loss.

Differential Equations

The null space of a differential operator corresponds to the solution space of the homogeneous differential equation, forming the foundation of many physical models.

Example: For the differential equation d²y/dx² + y = 0, the null space is spanned by "sin(x)" and "cos(x)", representing simple harmonic motion.

Economics and Input-Output Models

In economic input-output models, the null space can represent production scenarios that have no net impact on the economy, important for understanding economic equilibrium.

Network Analysis and Conservation Laws

In network flow problems, the null space often represents conservation principles (like Kirchhoff's laws in electrical circuits or conservation of mass in fluid networks).

Example: In electrical networks, the null space of the incidence matrix represents all possible current loops that satisfy Kirchhoff's Current Law.

Machine Learning and Data Analysis

In machine learning, understanding the null space helps identify redundant features and improve model robustness by recognizing which combinations of variables provide no additional information.

Example: In Principal Component Analysis (PCA), the null space of the covariance matrix corresponds to directions in the data with zero variance, which can be safely removed for dimensionality reduction.

Examples of Null Spaces

Example 1: Identity Matrix

For the identity matrix:

(100010001)\begin{pmatrix} 1 & 0 & 0 \\ 0 & 1 & 0 \\ 0 & 0 & 1 \end{pmatrix}

The null space contains only the zero vector because the system Ix = 0 has only the trivial solution x = 0.

Null(I)={0⃗}\text{Null}(I) = \{\vec{0}\}

This reflects that the identity transformation preserves all vectors, so only the zero vector maps to zero.

Example 2: Rank-Deficient Matrix

For the matrix:

(123246369)\begin{pmatrix} 1 & 2 & 3 \\ 2 & 4 & 6 \\ 3 & 6 & 9 \end{pmatrix}

This matrix has rank 1, and the null space has dimension 2. A basis for the null space is:

{(−210),(−301)}\left\{\begin{pmatrix} -2 \\ 1 \\ 0 \end{pmatrix}, \begin{pmatrix} -3 \\ 0 \\ 1 \end{pmatrix}\right\}

Any vector in the null space can be written as a linear combination of these basis vectors. Geometrically, the null space is a plane in R³.

This matrix represents a projection onto a line in 3D space, collapsing two dimensions to zero.

Example 3: Projection Matrix

For the projection matrix onto the xy-plane:

(100010000)\begin{pmatrix} 1 & 0 & 0 \\ 0 & 1 & 0 \\ 0 & 0 & 0 \end{pmatrix}

The null space is:

Null(P)=span{(001)}\text{Null}(P) = \text{span}\left\{\begin{pmatrix} 0 \\ 0 \\ 1 \end{pmatrix}\right\}

This is the z-axis, which makes geometric sense because all vectors along the z-axis are projected to the zero vector on the xy-plane.

Example 4: Zero Matrix

For the zero matrix:

(000000)\begin{pmatrix} 0 & 0 & 0 \\ 0 & 0 & 0 \end{pmatrix}

The null space is the entire domain R³, since every vector is mapped to zero.

Null(0)=R3\text{Null}(0) = \mathbb{R}^3

This represents complete information loss in the transformation.

Historical Context of the Null Space

The concept of the null space emerged from the work on solving systems of linear equations, dating back to ancient civilizations, but was formalized in the development of linear algebra:

Early Foundations (1800s)

The theoretical foundation for the null space began with Gaussian elimination, named after Carl Friedrich Gauss, though similar methods were used by Chinese mathematicians as early as 200 BCE.

Formal Development (Early 1900s)

The formal concept of the null space emerged with the development of vector spaces and linear transformations by mathematicians like Giuseppe Peano, David Hilbert, and Hermann Grassmann.

Modern Linear Algebra (Mid-1900s)

The term "kernel" (or null space) was standardized with the work of Emil Artin and Claude Chevalley in the 1940s, who helped establish the axiomatic approach to linear algebra we use today.

Computational Era (Late 1900s)

With the advent of computers, numerical methods for computing the null space became essential in applications ranging from engineering to economics, driving new algorithms for stable computation.

"The null space stands as one of the most fundamental concepts in linear algebra, bridging pure mathematics with countless practical applications across science and engineering."

Null Space and Fundamental Subspaces

The null space is one of the four fundamental subspaces associated with a matrix A, forming a complete characterization of the matrix as a linear transformation:

Column Space (Range)

The span of the columns of A, denoted Col(A). Represents the range of the linear transformation.

Col(A)⊂Rm\text{Col}(A) \subset \mathbb{R}^m

Dimension = rank(A)

Null Space (Kernel)

The set of vectors x such that Ax = 0, denoted Null(A). Represents the kernel of the linear transformation.

Null(A)⊂Rn\text{Null}(A) \subset \mathbb{R}^n

Dimension = n - rank(A)

Row Space

The span of the rows of A, denoted Row(A). Equals Col(AT).

Row(A)⊂Rn\text{Row}(A) \subset \mathbb{R}^n

Dimension = rank(A)

⊥ Null(A) (orthogonal complement)

Left Null Space

The set of vectors y such that yTA = 0, denoted Null(AT).

Null(AT)⊂Rm\text{Null}(A^T) \subset \mathbb{R}^m

Dimension = m - rank(A)

⊥ Col(A) (orthogonal complement)

Key Relationships

  • The row space and null space are orthogonal complements in Rn
  • The column space and left null space are orthogonal complements in Rm
  • The row space and column space have the same dimension (= rank(A))
  • The four subspaces completely characterize the action of A as a linear transformation

Visualizing the Null Space

The null space can be visualized geometrically in different ways, depending on its dimension:

Zero-Dimensional

When nullity = 0, the null space contains only the origin. For a 3×3 invertible matrix, this is a single point in R³.

Geometric meaning: The transformation maps no non-zero vectors to zero; all input information is preserved.

One-Dimensional

When nullity = 1, the null space is a line through the origin. For a 3×3 matrix with rank 2, this is a line in R³.

Geometric meaning: The transformation collapses one dimension, mapping all vectors along a particular direction to zero.

Two-Dimensional

When nullity = 2, the null space is a plane through the origin. For a 3×3 matrix with rank 1, this is a plane in R³.

Geometric meaning: The transformation collapses two dimensions, preserving only one dimension of information.

Full Dimensional

When nullity = n (the zero matrix), the null space is the entire space Rn.

Geometric meaning: The transformation collapses all dimensions, mapping every vector to zero.

"The geometry of the null space reveals what information is lost during a linear transformation."

⚠️For educational and informational purposes only. Verify with a qualified professional.

🧮 Fascinating Math Facts

📐

Null(A) is a subspace

⚡

dim = n − rank

👈 START HERE
⬅️Jump in and explore the concept!
AI