Scribd Logo
Upload

Welcome to Scribd!

  • 14 views

    Matrices

    Uploaded by

    Dakalo
    Matrices notes

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    Matrices

    Uploaded by

    Dakalo
    14 views8 pages
    Matrices notes

    Original Title

    matrices

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Matrices notes

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    14 views8 pages

    Matrices

    Uploaded by

    Dakalo
    Matrices notes

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 8

    Matrix

    calculations

    Walter
    Mudzimbabwe

    Operations

    Special
    matrices
    Matrix calculations
    Vector norms

    Matrix norms

    Walter Mudzimbabwe
    Matrix operations

    Matrix
    calculations
    Let A ∈ Rm×n then we write
    Walter
    Mudzimbabwe  
    a11 · · · a1n
    Operations  .. .. .. 
    Special
    A = [aij ] =  . . . 
    matrices
    am1 · · · amn
    Vector norms

    Matrix norms
    Addition: Rm×n × Rm×n −→ Rm×n : C = A + B where
    cij = aij + bij
    Scalar multiplication: R × Rm×n −→ Rm×n : C = αA where
    cij = αaij
    Multiplication: Rm×n × Rn×p −→ Rm×p : C = AB where
    Pn
    cij = aik bkj
    k=1
    Transpose: Rm×n −→ Rn×m : C = AT where cij = aji
    Special square matrices

    Matrix
    calculations
    Symmetric matrix: An A ∈ Rn×n is symmetric if AT = A.
    Walter Examples:
    Mudzimbabwe  
      4 0 1 10
    Operations 1 7 3
    7 4 5 , 
    0 −3 6 −2 .
    Special
    matrices 1 6 1 10 
    3 5 0
    Vector norms 10 −2 10 10
    Matrix norms
    Orthogonal matrix: An A ∈ Rn×n is orthogonal if AT A = In
    where In ∈ Rn×n is an identity matrix.
    The columns and rows are orthonormal ie if you take the dot
    product of any two columnns or rows, the product is 0.
    Examples:
     
        −1 0 0
    cos(θ) − sin(θ) 1 0
    , ,  0 −1 0
    sin(θ) cos(θ) 0 −1
    0 0 1
    Orthogonal matrices

    Matrix
    calculations

    Walter
    Mudzimbabwe

    Operations

    Special
    matrices

    Vector norms From AT A = In we have A−1 = AT which makes finding the


    Matrix norms inverse much easier (than doing lots of row operations, imagine
    if n = 1000!)
    Other Special non square matrices

    Matrix
    calculations

    Walter Diagonal matrix: An A ∈ Rm×n is a diagonal matrix if aij = 0


    Mudzimbabwe
    when i 6= j.
    Operations Notation: We write A =diag(α1 , α2 , · · · , αk ) where
    Special
    matrices
    k = min{m, n} then
    Vector norms

    Matrix norms
    A = [aij ] is diagonal and aii = αi for i = 1, 2, · · · , k

    Examples:
       
    7 0   4 0 0
    0 −5 , 3 0 0 0 7
    , 0 .
    0 6 0
    0 0 0 0 −13

    Not that this is not necessarily square.


    Vector norms

    Matrix
    calculations

    Walter A vector norm on Rn is a function f : Rn −→ R such that:


    Mudzimbabwe
    1 f (x) ≥ 0, ∀ x ∈ Rn
    Operations
    2 f (x + y) ≤ f (x) + f (y), ∀ x, y ∈ Rn
    Special
    matrices
    3 f (αx) = |α|f (x), ∀ α ∈ R, ∀ x ∈ Rn .
    Vector norms

    Matrix norms
    We denote such an f (x) by ||x||.
    Examples:
    Holder/p-norms: ||x||p = (|x1 |p + |x2 |p + · · · + |xn |p )1/p For
    example:
    1 ||x||1 = |x1 | + |x2 | + · · · + |xn |
    2 ||x||2 = (|x1 |2 + |x2 |2 + · · · + |xn |2 )1/2 = (xT x)1/2
    3 ||x||∞ = max |xi |
    i
    Vector norms

    Matrix
    calculations
    Exercise: Prove that the 2-norm is invariant under orthogonal
    Walter
    Mudzimbabwe transformations
    Solution: By a an orthogonal transformation of x, we mean Qx
    Operations

    Special
    where Q is an orthogonal matrix. So the question is asking you
    matrices to prove that ||Qx||2 = ||x||2 .
    Vector norms Now
    Matrix norms

    ||Qx||22 = (Qx)T Qx
    = xT QT Qx
    = xT x, since QT Q = In , because Q is orthogonal
    = ||x||22

    which implies ||Qx||2 = ||x||2 since the norm can not be


    negative.
    Matrix norms

    Matrix
    calculations

    Walter
    Mudzimbabwe
    A matrix norm on Rm×n is a function f : Rm×n −→ R such
    that:
    Operations

    Special
    1 f (A) ≥ 0, ∀ A ∈ Rm×n with f (A) = 0 iff A = 0
    matrices
    2 f (A + B) ≤ f (A) + f (B), ∀ A, B ∈ Rm×n
    Vector norms

    Matrix norms 3 f (αA) = |α|f (A), ∀ α ∈ R, ∀ A ∈ Rm×n .


    Again, we denote such an f (x) by ||x||.
    Examples:
    !1/2
    m P n
    |aij |2
    P
    Frobenius norm: ||A||F = .
    i j
    We can also define p-norms but they are not neccessary in this
    course.

    You might also like