click below
click below
Normal Size Small Size show me how
MATH 416 Midterm 3
| Term | Definition |
|---|---|
| Column rank (A) | Dimension of the span of the columns of A |
| Row rank (A) | Dimension of the span of the rows of A |
| Suppose A is an m-by-n matrix with column rank (A) and row rank (A) | Column rank (A) = row rank (A), and we call this rank (A) |
| Polynomials on operators | P(T) = a0 Iv + a1T + ... + amT^m. P(T) is a linear operator over the field F |
| Fundamental theorem of algebra | Every non-constant polynomial with complex coefficients has a root (also the unique factorization part, but it's not really necessary for this midterm) |
| Every operation on a finite-dimensional, non-zero, complex vector space has an eigenvalue | Not really an if/then or a definition |
| Upper triangular matrices | A matrix is called upper triangular if all the entries below the diagonal are equal to 0. |
| Equivalent conditions for upper triangular matrices | a) A is upper triangular b) T(vj) is in span(v1,...,vj) for each 0 < j < n+1 c) span(v1,...,vj) is invariant under T for each 0 < j < n+1 |
| Suppose V is a finite-dimensional vector space over C and T is a linear operator. | There exists a basis such that M(T) wrt that basis is upper triangular |
| A is an upper triangular matrix. A is invertible iff | A is an upper triangular matrix. All the diagonal entires are nonzero iff |
| Suppose T is a linear operator and M(T) is an upper triangular matrix representation. | Then the diagonal elements of M(T) are precisely the eigenvalues of T |
| Eigenspace | Suppose T is a linear operator and λ is in F. The eigenspace of T corresponding to λ (denoted by E( λ, T) ) is defined by null( T - λI ) |
| Suppose V is finite-dimensional and T is a linear operator. Suppose also that λ1,...,,λm are distinct eigenvalues of T. | Then E(λ1,T) + ... + E(λm,T) is a direct sum, and the sum of their dimensions is less than or equal to the dimension of V. |
| Diagonalizable | An operator T is called diagonalizable if the operator has a matrix representation giving a diagonal matrix wrt some basis of V. |
| Conditions equivalent to diagonalizability | a) T is diagonalizable b) V has a basis consisting of eigenvectors of T c) V = E(λ1,T) + ... + E(λm,T) (a direct sum) d) dim V = dim( E(λ1,T) ) + ... + dim( E(λm,T) ) |
| If T has dim(V) distinct eigenvalues | Then T is diagonalizable |
| Inner product | An inner product is a function that takes each ordered pair of two vectors from V to a field element, has the following properties: Positivity, definiteness ( <u1,u1>=0 iff u1=0), first slot additivity, first slot homogeneity, and conjugate symmetry |
| Inner product space | A vector space V with an inner product on V |
| Basic properties of inner products (a) | For each fixed u2 in V, the function that takes u1 to <u1,u2> is a linear map from V to C |
| Basic properties of inner products (b) | <0,u> = 0 = <u,0> for all u in V |
| Basic properties of inner products (c) | <u1,u2+w> = <u1,u2> + <u1,w> for all u1, u2, w in V |
| Basic properties of inner products (d) | <u1, λu2> = λ'<u1,u2> for λ in C and u1,u2 in V where λ' is the complex conjugate of λ |
| Norm | For u in V, the norm squared of u ( ||u|| ) is defined by ||u||^2 = <u,u> |
| Basic properties of norm | a) ||u|| = 0 iff u = 0 b) ||λu|| = |λ| ||u|| |
| Orthogonal | Two vectors u1,u2 in V are orthogonal if <u1,u2> = 0 (order does not matter) |
| Orthogonal properties | a) 0 is orthogonal to every vector in V b) 0 is the only vector in V that is orthogonal to itself |
| Pythagorean theorem | Suppose u1,u2 are orthogonal vectors. ||u1 + u2||^2 = ||u1||^2 + ||u2||^2 |
| Cauchy-schwarz inequality | Suppose u, v are in V. Then | <u, v> | ≤ ||u|| ||v|| with the equality holding iff one of the vectors is a scalar multiple of the other |
| Triangle inequality | Suppose u, v are in V. Then ||u+v|| ≤ ||u|| + ||v|| with the equality holding iff one of the vectors is a non-negative multiple of the other |
| Parallelogram equality | Suppose u, v are in V. Then ||u+v||^2 + ||u-v||^2 = 2(||u||^2+||v||^2) |
| Orthonormal basis | A list of vectors is called orthonormal if each of the vectors in the list has norm 1 and is orthogonal to all other vectors in the list |
| If e1,...,em is an orthonormal list of vectors in V, then ||a1e1+...+amem||^2 = | = |a1|^2 + ... + |am|^2 |
| Every orthonormal list of vectors is linearly independent | Not an if/then or a definition |
| Every orthonormal list of vectors in V with length dim(V) is an ONB of V | With ONB standing for orthonormal basis |
| Suppose e1,...,en is an ONB of V and v1 is in V | Then v1 = <v1,e1>e1 + ... + <v1,en>en and ||v1||^2 equals the sum of |<v1,ei>|^2 with 1≤i≤n |
| Gram-Schmidt procedure | e1 = v1/||v1|| and ej = vj - <vj,e1,>e1 - ... - <vj,ej-1>ej-1 / (norm of numerator) |
| Every finite-dimensional inner product space has an orthonormal basis | Not an if/then or a definition |
| Suppose V is finite-dimensional. Then every orthonormal list of vectors in V can be extended to an orthonormal basis of V | Not an if/then or a definition I mean sort of but still |
| Orthogonal complement | If U is a subset of V, then the orthogonal complement of U (denoted by U⊥) is the set of all vectors in V orthogonal to every vector in U |
| Properties of orthogonal complement | a) U⊥ is a subspace of V b) (0)⊥ = V c) V⊥ = (0) d) U (intersection) U⊥ is a subset of (0) e) U,W are subsets of V, U is a subset of W, then W⊥ is a subset of U⊥ |
| If U is a subspace of V, | Then V = U + U⊥ (a direct sum) and dim(U⊥) = dim(V) - dim(U) |
| Suppose U is a subset of V and V is a finite-dimensional vector space | (U⊥)⊥ = U |
| Orthogonal projection | U is a subspace of V. The orthogonal projection of V onto U (denoted Pu) is defined as follows: for v' in V, write v' = u' + w' where u' in U and w' in U⊥, Pu(v') = u' |
| Properties of Pu | a) Pu is a linear operator b) For u' in U, Pu(u') = u' c) For w' in U⊥, Pu(w') = 0 d) Pu^2 = Pu e) = range(Pu) = U f) null(Pu) = U⊥ |
| Suppose U is a subspace of V, V is finite-dimensional, v1 in V and u1 in U | Then ||v1 -Pu(v1)|| ≤ ||v1-u1|| with the equality achieved iff u1 = Pu(v1) |
| Adjoints | Suppose T is a linear map from V to W. The adjoint of T (denoted T*) from W to V is defined such that <Tv,w> = <v,T*w> |
| Properties of adjoints | 1) T* is a linear map 2) (S+T)* = S* + T* 3) (λT)* = λ'T* where λ' is the complex conjugate of λ 4) (T*)* = T 5) (Iv)* = Iv 6) (ST)* = T*S* |
| Relationships between T and T* | null(T*) = range(T) ⊥ I'm not writing the rest out but they follow similarly from there |
| Conjugate transpose | An m-by-n matrix that we can obtain by interchanging rows and columns and then taking the complex conjugate of each entry of a matrix of size n-by-m |
| Let T be a linear map from V to W, (e) be an ONB of V, (f) be an ONB of W | M(T*, (f), (e)) is the conjugate transpose of M(T, (e), (f)) |
| Self adjoint operators | T is called self-adjoint if T*=T |
| Every eigenvalue of a self-adjoint operator is real | Not an if/then or a definition |
| <Tv, v> = 0 for all v in V where V is a complex inner product space iff | T = 0 iff |
| T = T* iff | <Tv, v> is real for all v in V |
| If T = T* and <Tv, v> = 0 for all v in V | Then T = 0 |