4.2 Null Spaces, Column Spaces, and Linear Transformations Two useful subspaces of a vector space that turn up in applications are the null space and the column space of a matrix A representing a linear transformation. The Null Space of A Definition: The null space of an m x n matrix A is the set (Nul A) of all solutions to the homogeneous equation Ax = 0. So, Nul A = {x: x Rn and Ax = 0} Or, Nul A is the set of all x in Rn that get mapped into the 0 in Rm under x Ax. Theorem 4.2: The null space of and an m x n matrix A is a subspace of Rn. Equivalently, the set of all solutions to a system Ax = 0 of m homogeneous linear equations in n unknowns is a subspace of Rn. Proof: Nul A is a subset of Rn, so we show: a. 0 is in Nul A, b. that Nul A is closed under vector addition, c. Nul A is closed under scalar multiplication. 1 a. A0=0, so 0 is in NulA b. Let u, v be in Nul A. We need to show that A(u + v) = 0 A(u + v) = Au + Av By the definition of the null space of A, Au = 0 & Av = 0, so Au + Av =0+0 =0 So, Nul A is closed under vector addition. c. Let u be in Nul A and c a scalar. Then, by the definition of the null space of A, Au = 0. Acu = cAu = c0 =0 So, Nul A is closed under scalar multiplication. Since properties a, b, and c hold, Nul A is a subspace of Rn. QED 2 ►Given a matrix A, we can solve the equation Ax = 0 to get a specific description of Null A. Example: Let 3 6 6 3 9 A 6 12 13 0 3 Row reduce the augmented matrix corresponding to Ax = 0 to get 1 2 0 13 33 0 0 1 6 15 0 0 The free variables are x2 , x4 , x5 The solution is: x1 2 x2 13x4 33x5 x x2 2 x3 6 x4 15 x5 x x 4 4 x5 x5 So, any vector that satisfies Ax = 0 is a linear combination of three vectors: 3 2 13 33 1 0 0 x2 0 x5 6 x5 15 0 1 0 0 0 1 If we name these three vectors u, v, w respectively, Nul A = Span{u, v, w} Note: The set {u, v, w} found using this method is automatically linearly independent. 2 13 33 0 1 0 0 0 c1 0 c2 6 c3 15 0 0 1 0 0 0 0 1 0 Look at the second, fourth, and fifth entries to see that c1 = c2 = c3 = 0 Note: If Nul A ≠ {0}, the number of vectors in the spanning set of Nul A equals the number of free variables in Ax = 0. 4 The Column Space of A Definition: The column space of an m x n matrix A is the set (Col A) of all linear combinations of the columns of A. That is, if A = [a1 … an], Col A = Span{a1, …, an}. OR! Col A = {b: b = Ax for some x in Rn}. From this definition, Col A is a subset of Rm. Col A is also a subspace of Rm. Theorem 4.3: The column space of an m x n matrix A is a subspace of Rm. Proof: Col A = Span{a1, …, an}. By Th. 4.1, Span{a1, …, an} is a subspace of the space that contains {a1, …, an}, and {a1, …, an} Rm. Thus, Col A is a subspace of Rm. QED 5 Example: Find a matrix A such that W = Col A where x 2 y W 3 y : x, y R x y Write W as the set of all linear combinations of vectors. 1 2 x 0 y 3 1 1 Let these vectors be called u and v respectively, then W = Span{u, v}, so a matrix whose column space is W is: 1 2 A 0 3 1 1 Note: By Th. 1.4, The column space of an m x n matrix A is all of Rm iff the equation Ax = b has a solution for every b in Rm. 6 Differences Between Nul A and Col A: Example: Let 1 2 A 3 0 2 3 4 7 6 10 0 1 ►The column space of A is a subspace of R?. The columns of A have 4 entries, so they and linear combinations’s of them are in R4. ► The null space of A is a subspace of R?. Null A contains vectors that can be multiplied by this matrix, so they must be in R3. ►Find a non-zero vector in Col A. Any linear combination of the columns will do: 1 2 3 2 4 7 c1 c2 c3 3 6 10 0 0 1 7 ► Find a non-zero vector in Nul A. Solve the equ. Ax = 0, and pick a solution. The augmented matrix of Ax = 0 is 1 2 3 0 2 3 0 4 7 0 6 10 0 0 1 0 The reduced row echelon form is: 1 0 0 0 2 0 0 0 1 0 0 0 0 0 0 0 Solution: x1 2 x2 x2 is free x3 0 Since x2 is free, we can choose any value for x2, say x2 = 1: 8 x1 1 x x2 1 x3 0 Be sure to read page 232 for more contrasts! Summing Up: A subspace H of a vector space V is a subset of V that has three properties: a. The zero vector of V is in H. b. u and v H, u + v H. c. u H, and each scalar c R, cu H. Theorems 4.1, 4.2, and 4.3 are ways of proving that sets are subspaces without resorting to the definition. The null space of an m x n matrix A is a subspace of Rn. The column space of an m x n matrix A is a subspace of Rm. 9 Kernel and Range of a Linear Transformation Definition: A linear transformation T from a vector space V into a vector space W is a rule that assigns to each vector x in V a unique vector T(x) in W such that i T(u + v) = T(u) + T(v) ii T(cu) = cT(u) for all u, v in V and all scalars c R. Definition: the kernel (or null space) of T is the set of all vectors u in V such that T(u) = 0. Definition: The range of T is the set of all vectors in W of the form T(u) where u is in V. So, if T(x) = Ax, Col A is the range of T. 10