LINEAR-ALGEBRA ยท Unit 11 ยท Video 4 ยท Interactive Practice

One Equation, Four Subspaces

IKey Formulas

Formula Name Description
dimโกC(AT)=r\dim C(A^T) = r Row Space Dimension Equals the rank
dimโกN(A)=nโˆ’r\dim N(A) = n - r Rank-Nullity Theorem Columns minus rank
dimโกC(A)=r\dim C(A) = r Column Space Dimension Equals the rank
dimโกN(AT)=mโˆ’r\dim N(A^T) = m - r Left Null Space Dim Rows minus rank

For an mร—nm \times n matrix of rank rr: row space and null space are orthogonal complements in Rn\mathbb{R}^n; column space and left null space are orthogonal complements in Rm\mathbb{R}^m.

IIVisualization 1 โ€” Null-Space Membership

For A=(1111)A = \begin{pmatrix} 1 & 1 & 1 & 1 \end{pmatrix}, a vector lies in N(A)N(A) exactly when its four components sum to zero.

Because AA has rank 11 and n=4n = 4 columns, this null space has dimension nโˆ’r=3n - r = 3 โ€” the three basis vectors above span it.

IIIVisualization 2 โ€” Orthogonal Complements

For A=(ab)A = \begin{pmatrix} a & b \end{pmatrix}, the null space is the line through the origin perpendicular to the row space.

The complement of a line is a line in R2\mathbb{R}^2 (1+1=21+1=2), a plane in R3\mathbb{R}^3 (1+2=31+2=3), and a 33-space in the video's R4\mathbb{R}^4 (1+3=41+3=4).

IVVisualization 3 โ€” The Dimension Balance

For any mร—nm \times n matrix of rank rr, the four subspace dimensions always satisfy r+(nโˆ’r)=nr+(n-r)=n and r+(mโˆ’r)=mr+(m-r)=m.

Rank can't exceed minโก(m,n)\min(m,n); push rr past it and it stays capped. At r=0r=0 the whole space is null space, and the dimensions still sum correctly.

VQuiz Questions

Question 1

The matrix A=(1111)A = \begin{pmatrix} 1 & 1 & 1 & 1 \end{pmatrix} is a 1ร—41 \times 4 matrix with rank 1.

What is the dimension of the null space N(A)N(A)?

โœ… Correct! By rank-nullity: dim N(A) = n โˆ’ r = 4 โˆ’ 1 = 3.

โŒ Not quite. Use the rank-nullity theorem: dim N(A) = n โˆ’ rank(A). Here n = 4 and rank = 1.

Show solution

Solution:

By the rank-nullity theorem: dimโกN(A)=nโˆ’r=4โˆ’1=3\dim N(A) = n - r = 4 - 1 = 3

The matrix has n=4n = 4 columns and rank r=1r = 1 (one pivot), so the null space has dimension 4โˆ’1=34 - 1 = 3.

The three basis vectors are: (โˆ’1,1,0,0)(-1, 1, 0, 0), (โˆ’1,0,1,0)(-1, 0, 1, 0), (โˆ’1,0,0,1)(-1, 0, 0, 1).

Question 2

Which of the following vectors is in the null space of A=(1111)A = \begin{pmatrix} 1 & 1 & 1 & 1 \end{pmatrix}?

โœ… Correct! The components sum to zero: 1 + (โˆ’1) + 0 + 0 = 0.

โŒ Not quite. A vector is in N(A) when its components sum to zero. Try adding up the components of each option.

Show solution

Solution:

A vector v\mathbf{v} is in N(A)N(A) if and only if v1+v2+v3+v4=0v_1 + v_2 + v_3 + v_4 = 0. Check each:

  • (1,1,1,1)(1, 1, 1, 1): 1+1+1+1=4โ‰ 01 + 1 + 1 + 1 = 4 \neq 0 โœ—
  • (1,โˆ’1,0,0)(1, -1, 0, 0): 1+(โˆ’1)+0+0=01 + (-1) + 0 + 0 = 0 โœ“
  • (โˆ’1,1,1,1)(-1, 1, 1, 1): โˆ’1+1+1+1=2โ‰ 0-1 + 1 + 1 + 1 = 2 \neq 0 โœ—
  • (0,0,0,1)(0, 0, 0, 1): 0+0+0+1=1โ‰ 00 + 0 + 0 + 1 = 1 \neq 0 โœ—

Only (1,โˆ’1,0,0)(1, -1, 0, 0) has components summing to zero.

Question 3

A matrix BB has size 3ร—73 \times 7 with rank 2.

What is the dimension of the null space N(B)N(B)?

โœ… Correct! dim N(B) = n โˆ’ r = 7 โˆ’ 2 = 5.

โŒ Not quite. Remember: dim N(B) = (number of columns) โˆ’ rank. Be careful not to use the number of rows!

Show solution

Solution:

By the rank-nullity theorem: dimโกN(B)=nโˆ’r=7โˆ’2=5\dim N(B) = n - r = 7 - 2 = 5

Here n=7n = 7 (columns) and r=2r = 2 (rank), so the null space has dimension 7โˆ’2=57 - 2 = 5.

Common mistakes:

  • Using mm instead of nn: 3โˆ’2=13 - 2 = 1 gives the left null space dimension, not the null space.
  • Answering with the rank itself (r=2r = 2) confuses rank with nullity.

Question 4

True or False: For a 1ร—n1 \times n matrix of rank 1, the left null space N(AT)N(A^T) always equals {0}\{\mathbf{0}\} (contains only the zero vector).

โœ… Correct! dim N(Aแต€) = m โˆ’ r = 1 โˆ’ 1 = 0, so the left null space is just {0}.

โŒ Not quite. Compute dim N(Aแต€) = m โˆ’ r. What do you get when m = 1 and r = 1?

Show solution

Solution: True

The left null space has dimension mโˆ’rm - r.

For a 1ร—n1 \times n matrix of rank 1: m=1m = 1 and r=1r = 1, so: dimโกN(AT)=mโˆ’r=1โˆ’1=0\dim N(A^T) = m - r = 1 - 1 = 0

A subspace of dimension 0 contains only the zero vector: N(AT)={0}N(A^T) = \{\mathbf{0}\}.

This is exactly what we saw in the video: the equation ATy=0A^T y = 0 with ATA^T being a column of ones forces y=0y = 0.

Solved: 0 / 4