(a). W contains the zero vector in R n. (b). W is closed under addition. (c). W is closed under scalar multiplication.

Transcription

1 . Subspaces of R n Bases and Linear Independence Definition. Subspaces of R n A subset W of R n is called a subspace of R n if it has the following properties: (a). W contains the zero vector in R n. (b). W is closed under addition. (c). W is closed under scalar multiplication. Fact.. If T is a linear transformation from R n to R m, then ker(t ) is a subspace of R n im(t ) is a subspace of R m

2 Example. Is W = {[ x y ] R : x, y } a subspace of R? See Figure,. Example. Is W = {[ x y ] R : xy } a subspace of R? See Figure, 4. Example. Show that the only subspaces of R are: { }, any lines through the origin, and R itself. Similarly, the only subspaces of R are: { }, any lines through the origin, any planes through, and R itself.

3 Solution Suppose W is a subspace of R that is neither the set { } nor a line through the origin. We have to show W = R. Pick a nonzero vector v in W. (We can find such a vector, since W is not { }.) The subspace W contains the line L spanned by v, but W does not equal L. Therefore, we can find a vector v in W that is not on L (See Figure 5). Using a parallelogram, we can express any vector v in R as a linear combination of v and v. Therefore, v is contained in W (Since W is closed under linear combinations). This shows that W = R, as claimed.

4 A plane E in R is usually described either by x + x + x = or by giving E parametrically, as the span of two vectors, for example, and. In other words, E is described either as ker[ ] or im 4

5 Similarly, a line L in R may be described either parametrically, as the span of the vector or by two linear equations Therfore L = im x x x = x x + x = = ker [ ] A subspace of R n is uaually presented either as the solution set of a homogeneous linear system (as a kernel) or as the span of some vectors (as an image). Any subspace of R n can be represented as the image of a matrix. 5

6 Bases and Linear Independence Example. Consider the matrix A = 4 Find vectors v, v,, v m in R that span the image of A. What is the smallest number of vectors needed to span the image of A? Solution We know from Fact.. that the image of A spanned by the columns of A, v =, v =, v =, v 4 = 4 6

7 Figure 6 show that we need only v and v to span the image of A. Since v = v and v 4 = v + v, the vectors v and v 4 are redundant; that is, they are linear combinations of v and v : im(a) = span( v, v, v, v 4 ) = span( v, v ). The image of A can be spanned by two vectors, but not by one vectors alone. 7

8 Definition. Linear independence; basis Consider a sequence v,..., v m of vectors in a subspace V of R n. The vectors v,..., v m are called linearly independent if nono of them is a linear combination of the others. We say that the vectors v,..., v m form a basis of V if they span V and are linearly independent. See last example. The vectors v, v, v, v 4 span V = im(a) but they are linearly dependent, because v 4 = v + v. Therefore, they do not form a basis of V. The vectors v, v, on the other hand, do span V and are linearly independent. 8

9 Definition. Linear relations Consider the vectors v,..., v m in R n. An equation of the form c v + c v c m v m = is called a (linear) relation among the vectors v i. There is always the trievial relation, with c = c = = c m =. Nontrivial relations may or may not exist among the vectors v,..., v m in R n. 9

10 Fact..5 The vectors v,..., v m in R n are linearly dependent if (and only if) there are nontrivial relations among them. Proof If one of the v i s a linear combination of the others, v i = c v + +c i v i +c i+ v i c m v m then we can find a nontrivial relation by subtracting v i from both sides of the equations: c v + +c i v i v i +c i+ v i c m v m = Conversely, if there is a nontrivial relation c v + + c i v i c m v m = then we can solve for v i and express v i as a linear combination of the other vectors.

11 Example. Determine whether the following vectors are linearly independent 4 5, , 5 7, Solution TO find the relations among these vectors, we have to solve the vector equation c c c c =

12 or c c c c 4 = In other words, we have to find the kernal of A. To do so, we compute rref(a). Using technology, we find that This shows the kernel of A is { }, because there is a leading in each column of rref(a). There is only the trivial relation among the four vectors and they are therefore linearly independent.

13 Fact..6 The vectors v,..., v m in R n are linearly independent if (and only if) ker or, equivalently, of rank v v... v m v v... v m = { } = m This condition implies that m n.

14 Fact..7 Consider the vectors v,..., v m in a subspace V of R n. The vectors v i are a basis of V if (and only if) every vector v in V can be expressed uniquely as a linear combination of the vectors v i. Proof Suppose vectors v i are a basis of V, and consider a vector v in V. Since the basis vectors span V, the vector v can be written as a linear combination of the v i. We have to demonstrate that this representation is unique. If there are two representations: v = c v + c v c m v m = d v + d v d m v m By subtraction, we find = (c d ) v + (c d ) v (c m d m ) v m 4

15 Since the v i are linearly independent, c i d i =, or c i = d i, for all i., suppose that each vector in V can be expressed uniquely as a linear combination of the vectors v i. Clearly, the v i. span V. The zero vector can be expressed uniquely as a linear combination of the v i, namely, as = v + v v m This means there is only the trivial relation among the v i : they are linearly independent.

16 See Figure 7. The vectors v, v, v, v 4 do not form a basis of E, since every vector in E can be expressed in more than one way as a linear combination of the v i. For example, but also v 4 = v + v + v + v 4 v 4 = v + v + v + v 4. Homework.:, 5, 9, 7, 8, 9, 9,, 9 5