Module B - Eigenvalue and Singular Value Decompositions

Module B - Eigenvalue and Singular Value Decompositions

GLOBAL INITIATIVE OF ACADEMIC NETWORKS Ivan Slapniˇcar MODERN APPLICATIONS OF NUMERICAL LINEAR ALGEBRA METHODS Module B - Eigenvalue and Singular Value Decompositions IIT INDORE, 2016 GLOBAL INITIATIVE OF ACADEMIC NETWORKS Ivan Slapniˇcar MODERN APPLICATIONS OF NUMERICAL LINEAR ALGEBRA METHODS Module B - Eigenvalue and Singular Value Decompositions https://github.com/ivanslapnicar/GIAN-Applied-NLA-Course Cover photo: TU Berlin, Institut f¨urMathematik IIT INDORE, 2016 Contents 1 Eigenvalue Decomposition - Definitions and Facts5 1.1 Prerequisites......................................5 1.2 Competences......................................5 1.3 Selected references...................................5 1.4 General matrices....................................5 1.4.1 Definitions...................................5 1.4.2 Facts.......................................6 1.4.3 Examples....................................7 1.4.4 Example.....................................9 1.4.5 Example..................................... 10 1.4.6 Example - Circulant matrix.......................... 12 1.5 Hermitian and real symmetric matrices........................ 14 1.5.1 Definitions................................... 14 1.5.2 Facts....................................... 14 1.5.3 Example - Hermitian matrix......................... 16 1.5.4 Example - real symmetric matrix....................... 21 1.6 Positive definite matrices............................... 22 1.6.1 Definitions................................... 22 1.6.2 Facts....................................... 22 1.6.3 Example - Positive definite matrix...................... 23 1.6.4 Example - Positive semidefinite matrix.................... 24 1.6.5 Example - Covariance and corellation matrices............... 25 2 Eigenvalue Decomposition - Perturbation Theory 28 2.1 Prerequisites...................................... 28 2.2 Competences...................................... 28 2.3 Norms.......................................... 28 2.3.1 Definitions................................... 28 2.3.2 Examples.................................... 29 2.3.3 Facts....................................... 29 2.4 Errors and condition numbers............................. 32 2.4.1 Definitions................................... 32 2.5 Peturbation bounds.................................. 32 2.5.1 Definitions................................... 32 2.5.2 Facts....................................... 33 2.5.3 Example - Nondiagonalizable matrix..................... 34 2.5.4 Example - Jordan form............................ 36 2.5.5 Example - Normal matrix........................... 40 2.5.6 Example - Hermitian matrix......................... 41 3 Symmetric Eigenvalue Decomposition - Algorithms and Error Analysis 47 3.1 Prerequisites...................................... 47 3.2 Competences...................................... 47 3.3 Backward error and stability............................. 47 1 3.3.1 Definitions................................... 47 3.4 Basic methods..................................... 47 3.4.1 Definitions................................... 47 3.4.2 Facts....................................... 48 3.4.3 Examples.................................... 49 3.5 Tridiagonalization................................... 52 3.5.1 Facts....................................... 52 3.6 Tridiagonal QR method................................ 57 3.6.1 Definition.................................... 57 3.6.2 Facts....................................... 57 3.6.3 Examples.................................... 58 3.6.4 Computing the eigenvectors.......................... 60 3.6.5 Symmetric QR method............................ 63 4 Symmetric Eigenvalue Decomposition - Algorithms for Tridiagonal Matrices 65 4.1 Prerequisites...................................... 65 4.2 Competences...................................... 65 4.3 Bisection and inverse iteration............................. 65 4.3.1 Facts....................................... 65 4.4 Divide-and-conquer................................... 67 4.4.1 Facts....................................... 67 4.5 MRRR.......................................... 70 5 Symmetric Eigenvalue Decomposition - Jacobi Method and High Relative Accuracy 72 5.1 Prerequisites...................................... 72 5.2 Competences...................................... 72 5.3 Jacobi method..................................... 72 5.3.1 Definitions................................... 72 5.3.2 Facts....................................... 73 5.3.3 Examples.................................... 73 5.4 Relative perturbation theory............................. 76 5.4.1 Definition.................................... 76 5.4.2 Facts....................................... 76 5.4.3 Example - Scaled matrix............................ 77 5.5 Indefinite matrices................................... 79 5.5.1 Definition.................................... 79 5.5.2 Facts....................................... 79 5.6 # Symmetric Eigenvalue Decomposition - Lanczos Method............ 79 5.7 Prerequisites...................................... 80 5.8 Competences...................................... 80 5.9 Lanczos method.................................... 80 5.9.1 Definitions................................... 80 5.9.2 Facts....................................... 80 5.9.3 Examples.................................... 81 2 5.9.4 Operator version................................ 90 5.9.5 Sparse matrices................................. 93 6 Singular Value Decomposition - Definitions and Facts 95 6.1 Prerequisites...................................... 95 6.2 Competences...................................... 95 6.3 Selected references................................... 95 6.4 Singular value problems................................ 95 6.4.1 Definitions................................... 95 6.4.2 Facts....................................... 95 6.4.3 Example - Symbolic computation....................... 97 6.4.4 Example - Random complex matrix..................... 98 6.4.5 Example - Random real matrix........................ 100 6.5 Matrix approximation................................. 102 7 Singular Value Decomposition - Perturbation Theory 104 7.1 Prerequisites...................................... 104 7.2 Competences...................................... 104 7.3 Peturbation bounds.................................. 104 7.3.1 Definitions................................... 104 7.3.2 Facts....................................... 104 7.3.3 Example..................................... 105 7.4 Relative perturbation theory............................. 108 7.4.1 Definitions................................... 108 7.4.2 Facts....................................... 108 7.4.3 Example - Bidiagonal matrix......................... 109 8 Singular Value Decomposition - Algorithms and Error Analysis 113 8.1 Prerequisites...................................... 113 8.2 Competences...................................... 113 8.3 Basics.......................................... 113 8.3.1 Definitions................................... 113 8.3.2 Facts....................................... 113 8.4 Bidiagonalization.................................... 114 8.4.1 Facts....................................... 114 8.5 Bidiagonal QR method................................. 118 8.5.1 Facts....................................... 118 8.5.2 Examples.................................... 119 8.6 QR method....................................... 122 9 Singular Value Decomposition - Jacobi and Lanczos Methods 125 9.1 Prerequisites...................................... 125 9.2 Competences...................................... 125 9.3 One-sided Jacobi method............................... 125 9.3.1 Definition.................................... 125 9.3.2 Facts....................................... 125 3 9.3.3 Example - Standard matrix.......................... 126 9.3.4 Example - Strongly scaled matrix....................... 128 9.4 Lanczos method.................................... 132 9.4.1 Example - Large matrix............................ 135 9.4.2 Example - Very large sparse matrix..................... 135 10 Algorithms for Structured Matrices 138 10.1 Prerequisites...................................... 138 10.2 Competences...................................... 138 10.3 Rank revealing decompositions............................ 138 10.3.1 Definitions................................... 138 10.3.2 Facts....................................... 138 10.3.3 Example - Positive definite matrix...................... 140 10.3.4 Example - Hilbert matrix........................... 142 10.4 Symmetric arrowhead and DPR1 matrices...................... 149 10.4.1 Definitions................................... 149 10.4.2 Facts on arrowhead matrices......................... 149 10.4.3 Example - Random arrowhead matrix.................... 150 10.4.4 Example - Numerically demanding matrix.................. 152 10.4.5 Facts on DPR1 matrices............................ 152 10.4.6 Example - Random DPR1 matrix....................... 153 10.4.7 Example - Numerically demanding matrix.................. 153 11 Updating the SVD 155 11.1 Prerequisites...................................... 155 11.2 Competences...................................... 155 11.3 Facts........................................... 155 11.3.1 Example - Adding row to a tall matrix.................... 156 11.3.2 Example - Adding row to a flat matrix.................... 158 11.3.3 Example - Adding columns.......................... 158 11.3.4 Example - Updating a

View Full Text

Details

  • File Type
    pdf
  • Upload Time
    -
  • Content Languages
    English
  • Upload User
    Anonymous/Not logged-in
  • File Pages
    165 Page
  • File Size
    -

Download

Channel Download Status
Express Download Enable

Copyright

We respect the copyrights and intellectual property rights of all users. All uploaded documents are either original works of the uploader or authorized works of the rightful owners.

  • Not to be reproduced or distributed without explicit permission.
  • Not used for commercial purposes outside of approved use cases.
  • Not used to infringe on the rights of the original creators.
  • If you believe any content infringes your copyright, please contact us immediately.

Support

For help with questions, suggestions, or problems, please contact us