- Class room: EECS131
- Class time: W3W4F4
- Office hour: by appointment(cherung@ cs. nthu.edu.tw)
- Lecturer: Che-Rung Lee
- Reference books
- Y. Saad, "Iterative methods for sparse linear systems (2nd edition)", link
- Y. Saad, "Numerical Methods for Large Eigenvalue Problems - 2nd Edition", link
- G. W. Stewart, "Matrix Algorithms, Volume II: Eigensystems", (E-book in library)
- G. W. Stewart, "Matrix Algorithms, Volume I: Basic Decompositions". (E-book in library)
- James Demmel, "Applied Numerical Linear Algebra" (E-book in library)
- Tim Davis, "Direct Methods for Sparse Linear Systems" (E-book in library)
- Gene H. Golub and Charles F. Van Loan, "Matrix Computation"

- Reference links

- Homework: 30%
- Class note: 30%
- Project: 40%
- Pick one topic and one (or several) paper(s) to read and do some experiments.
- Discuss the project with me bi-weekly. Schedule
- Write a report (4-6 pages).

Date | Topic | Reference | Note |
---|---|---|---|

0 | Introduction | Top 10 Algorithms in the 20th Century | |

Matrix multiplication | |||

1 | Basic and block formulation |
BLAS LAPACK CLAPACK Prof. Demmel's talk | note01 |

2 | Parallel matrix multiplication | note02 | |

3 | Strassen algorithm |
note03_1, note03_2 | |

4 | Fast Fourier Transformation | note04 | |

Sparse matrices |
Matrix Market Sparse Matrix Collection | hw1.pdf | |

Basic matrix decomposition | |||

5 | LU decomposition and Cholesky decomposition | note05 | |

6 | QR decomposition: Gram-Schmidt process | note06 | |

QR decomposition: Householder reflection | |||

7 | QR decomposition: Green's rotation | note07 | |

Block version decompositions and parallelization | |||

Eigenvalue problems | |||

8 | Eigenvalue problems: Power methods | note08 | |

9 | Eigenvalue problems: Orthogonal iteration | note09 | |

10 | Eigenvalue problems: QR method | note10 | |

11 | Symmtric eigenvalue problems and SVD | note11 | |

Krylov subspace methods | |||

12 | Definition and theorems | note12 | |

13 | GMRES | note13 | |

14 | CG | cg |