Author| Rickyの水果摊
Time | 2022.9.12
Lecture 6: Column space and nullspace
Lecture Info
-
Instructor: Prof. Gilbert Strang
-
Course Number: 18.06
-
Topics: Linear Algebra
-
Official Lecture Resource: Resource Index of Linear Algebra
Excellent Notes on GitHub
There are some classic, excellent notes from other authors on GitHub, wihch I highly recommend you to star ⭐️ and read 📖
notes-linear-algebra (A systematic notes written in Chinese)
The-Art-of-Linear-Algebra (Focus on visualization of important concept of Linear Algebra)
Video Link
Key Points
- column space
- nullspace
Active Recall Questions
- Suppose we have subspaces P , Q P,Q P,Q , if P ∩ Q P \cap Q P∩Q / P ∪ Q P \cup Q P∪Q a subspace?
- What is the mathematical meaning of A x Ax Ax ?
- Does A x = b Ax=b Ax=b always have solutions for every b ? If the answer is no, which b’s allow A x = b Ax=b Ax=b to be solvable ? (From the perspective of C ( A ) C(A) C(A))
- What’s the definiton of the nullspace N ( A ) N(A) N(A) ?
- Suppose A m ∗ n A_{m*n} Am∗n , why N ( A ) N(A) N(A) is a subspace of R n \mathbb{R}^n Rn ? (Hint: rules that a subspace have to follow)
- What are the similarities and differences between C ( A ) , N ( A ) C(A), N(A) C(A),N(A) ? (Hint: focus, dimension, the way they span, …)
- Suppose all the solutions ( x s xs xs) to A x = b Ax=b Ax=b ( b b b is not zero vector) form a new space N ( A ) ′ N(A)' N(A)′, if N ( A ) ′ N(A)' N(A)′ a subspace? (Hint: compare to the N ( A ) N(A) N(A) that A x = 0 Ax=0 Ax=0 form)
Answer
-
P ∩ Q P \cap Q P∩Q is a subspace, P ∪ Q P \cup Q P∪Q is not a subspace
-
A x Ax Ax are all the linear combinations of columns of A A A, which form C ( A ) C(A) C(A) (Some columns of A contribute nothing to C ( A ) C(A) C(A), we will discuss them in the next lecture)
-
No. A x = b Ax=b Ax=b is solvable only when b b b is in C ( A ) C(A) C(A)
-
The nullspace N ( A ) N(A) N(A) consists of all solutions to A x = 0 Ax=0 Ax=0.
-
If we want to check if a space is a subspace, we need to check it from the definition of subspace
First, we take any two vectors x 1 , x 2 x_1,x_2 x1,x2 in the nullspace, they all satisfy A x = 0 Ax=0 Ax=0, we need to check if x 3 = c 1 ∗ x 1 + c 2 ∗ x 2 x_3=c_1*x_1+c_2*x_2 x3=c1∗x1+c2∗x2 is still in the nullspace.
If we can prove A x 3 = 0 Ax_3=0 Ax3=0, then we know that x 3 x_3 x3 is in N ( A ) N(A) N(A), and here is the derivation: A x 3 = A ( c 1 ∗ x 1 + c 2 ∗ x 2 ) = c 1 ∗ A x 1 + c 2 ∗ A x 2 = c 1 ∗ 0 + c 2 ∗ 0 = 0 Ax_3=A(c_1*x_1+c_2*x_2)=c_1*Ax_1+c_2*Ax_2=c_1*0+c_2*0=0 Ax3=A(c1∗x1+c2∗x2)=c1∗Ax1+c2∗Ax2=c1∗0+c2∗0=0
That means x 3 x_3 x3 is still in N ( A ) N(A) N(A). Hence N ( A ) N(A) N(A) is a subspace
-
Here are their similarites and differences:
- C ( A ) C(A) C(A) focus on A A A, N ( A ) N(A) N(A) focus on x x x.
- dim of C ( A ) C(A) C(A) is r a n k ( A ) rank(A) rank(A), dim of N ( A ) N(A) N(A) is ( n − r a n k ( A ) ) (n-rank(A)) (n−rank(A))
- C ( A ) C(A) C(A) is a subspace of R m \mathbb {R}^m Rm, N ( A ) N(A) N(A) is a subspace of R n \mathbb {R}^n Rn
- C ( A ) C(A) C(A) is formed by the pivot columns of A A A, N ( A ) N(A) N(A) is formed by all the solutions to A x = 0 Ax=0 Ax=0 (We will learn how to determine pivot columns and solutions to A x = 0 Ax=0 Ax=0 in the next lecture)
-
This derivation is similar to Q5.
First, we take any two vectors x 1 , x 2 x_1,x_2 x1,x2 in N ( A ) ′ N(A)' N(A)′, they all satisfy A x = b Ax=b Ax=b, we need to check if x 3 = c 1 ∗ x 1 + c 2 ∗ x 2 x_3=c_1*x_1+c_2*x_2 x3=c1∗x1+c2∗x2 is still in N ( A ) ′ N(A)' N(A)′
If we can prove A x 3 = b Ax_3=b Ax3=b, then we know that x 3 x_3 x3 is in N ( A ) ′ N(A)' N(A)′, however A x 3 = A ( c 1 ∗ x 1 + c 2 ∗ x 2 ) = c 1 ∗ A x 1 + c 2 ∗ A x 2 = c 1 ∗ b + c 2 ∗ b ≠ b Ax_3=A(c_1*x_1+c_2*x_2)=c_1*Ax_1+c_2*Ax_2=c_1*b+c_2*b \ne b Ax3=A(c1∗x1+c2∗x2)=c1∗Ax1+c2∗Ax2=c1∗b+c2∗b=b
That means x 3 x_3 x3 is not in N ( A ) ′ N(A)' N(A)′. Hence N ( A ) ′ N(A)' N(A)′ is not a subspace