Properties of Linear Transformations Theorem: If V and W are vector spaces, T: V W is a linear transformation, and u and v are vectors in V, then T has the following properties: 1) T(0) = 0; 2) T(- v) = - T(v); 3) T( u - v) = T(u) - T(v). Linear transformations preserve linear combinations: Let T: V W be a linear transformation, and let v1 , v 2 ,..., v n be vectors in V. If v c1 v 1 c 2 v 2 ... c n v n . Then, T ( v) c1T ( v 1 ) c 2 T ( v 2 ) ... c nT ( v n ) Thus, A linear transformation defined on V is completely determined by its values on a set of basis vectors for V: Example: Consider the two vectors v 1 = (2, 4) and v 2 = (4, 2) 2 . These vectors are a basis for 2 . Suppose we want to find the linear transformation T: 2 2 such that, T ( v1 ) w1 (3, 5) and T ( v 2 ) w 2 (3, 2) We know that there is a 2 2 matrix A such that T is defined by T ( v ) Av . Now Av1 v 2 Av1 Av 2 T ( v1 ) T ( v 2 ) w1 w 2 Thus, 2 A 4 4 3 3 . 2 5 2 So, 3 2 3 A 5 2 4 4 2 1 .5 1.5 .5 2 Hence, .5 x .5 T ( x, y ) (.5 x .5 y, - 1.5x 2y) - 1.5 2 y Definition: The linear transformation T : V W is one-to-one if no two different vectors have the same image in W. Definition: The transformation T : V W is onto if every vector in W is the image of at least one vector in V. Examples: 1. The transformation T : 2 3 defined by T ( x, y ) ( x, y, 1) is one-to-one but is not onto. Why is transformation not onto? 2. The transformation T : 3 1 defined by T ( x, y, z ) x is onto but is not one-toone. Why is the transformation not one-to-one? Definition: The linear transformation T : V W is an isomorphism if it is both one-toone and onto. In this case, given w W , there exists exactly one vector v V , such that T ( v ) w. Definition: The vector space V is isomorphic to the vector space W if there exists an isomorphism T : V W . Note: If T : V W is an isomorphism, then T 1 : W V is defined because T is one-toone and onto. Also, T 1 : W V is a linear transformation and therefore an isomorphism. Hence, the vector space W is isomorphic to the vector space V. In such a case we say that the vector spaces V and W are isomorphic. Exercise: Show that an n-dimensional vector space is isomorphic to n (This implies that any two n-dimensional vector spaces are isomorphic). Definition: Let T : V W be a linear transformation. The set of all vectors v in V such that T(v) = 0 is called the kernel of the linear transformation T, and is denoted by ker(T). (Note: ker(T) is a subspace of V) Exercise: Show that the linear transformation T : V W is one-to-one if and only if ker(T) ={0}(the zero subspace of V). Definition: Let T : V W be a linear transformation. The set of all vectors in W that are images under T of vectors in V, is called the range of the linear transformation T and is denoted by range(T). (Note: range(T) is a subspace of W) Result: The mapping T : V W is onto if and only if range(T) = W. Note: It the range(T) is finite dimensional, then its dimension is called the rank of T, denoted by rank(T). Consider the linear transformation T : n m given by T ( x) Ax , where A is m n , and x n . Then 1. ker (T) = Null(A), i.e., ker (T) is the solution space of the homogeneous linear system Ax = 0. 2. range(T) = Col(A). 3. rank(T) = dim Col(A) = rank(A). 4. rank(T) + Null(A) = n Example: The 3 5 matrix A of a linear transformation 8 1 2 2 A 2 3 1 3 1 1 1 4 1 rref (A) 0 0 T : 5 3 is given by 1 11 4 0 0 16/3 1 0 0 1 9/ 4 11 / 12 3 4 5 The pivot columns are the first, second, and third. Hence a basis for Col(A) = {(1, 2, 1), (-2, 3, 1), (-2, -1, 1)}, i.e. rank(T) = 3. To determine ker(T): Solve 16 u 3v 0 3 9 y u 4v 0 4 11 z u 5v 0 12 x Hence, x 16 a 3b 3 9 x 16 / 3 3 a 4b y 9 / 4 4 4 11 z a 5b z 11 / 12 a 5 b 12 u 1 0 ua v 0 1 vb y Hence a basis for ker(T) = {(-16/3, 9/4, -11/12, 1, 0), (3, -4, 5, 0, 1)} and dim ker(T) = 2.