Applied Numerical Methods, part 2

Course news, autumn 2010

• Room for the re-examination: 4523.
• The re-examination paper has been scheduled now. It will be given on April 8, 2011, 15-18. Those who like to attend shall drop me an email as soon as possible.
• The course evaluation form is online now.
• The rooms for the examination have changed: It is now Q1 and Q22.
• An excerpt of the Elden paper can be downloaded for lab 3 here. Get it!
• Course lecture notes which you can buy at the students' expedition: It is only the cover which is in Swedish. The text is in English!
• Lecture notes on rounding error analysis: get it!
• Here you will find some sample examinations: exam0601.pdf, exam0710.pdf.
• When using matlab's IDE on Solaris machines the caret sign (^) is not immediately available. For making it available to matlab issue the command

   xmodmap -e "keycode 55 = SunFA_Diaeresis asciicircum asciitilde"
  

  before you start matlab. Then, the caret key should work fine.
• The course starts in lecture hall D2 on Monday October 25, 2010 at 10.00 o'clock.
• If you are not yet registered for the course please contact your "studievägledare" for doing that.

Instructors

Michael Hanke, room D4529, phone 08-7906278: Lectures, Examiner
Jennifer Grünig, room D1524, phone 08-790 6927, jgrunig@csc.kth.se: lab sessions, advisor.

Plan for lectures 2010

Lectures, Föreläsningar, F1-F8 are given in various lecture halls around campus. Look carefully at your maps!
Lab consultations, L1-L7, are held in room 4523 (exception: L3) in the main building D, eastern wing 5th floor. This room is not equipped with computers!

Week  When         Where What  Contents (Reading)
 43   25/10  10-12  D2   Lect  Intr, Foundations of error analysis (L1, D1.2-3)
      26/10  10-12  4523 Lab cons
      28/10  10-12  H1   Lect  Error anal (cont), Gaussian eliminination (D2.1-4)
 44    2/11  10-12  4523 Lab cons
       4/11  10-12  H1   Lect  Direct methods for sparse matrices (L2, D2.7)
 45    9/11  10-12  1439 Lab cons
      10/11  10-12  L1   Lect  Linear lest squares, SVD (L3, D3.1,2,5)
 46   15/11  10-12  B1   Lect  SVD, eigenvalue problems (L4, D4)
      16/11  10-12  4523 Lab cons
 47   22/11  10-12  K1   Lect  Eigenvalue problems (cont)
      23/10  10-12  4523 Lab cons
 48   29/11  13-15  V2   Lect  Large-scale problems: Iterative methods (L5, D6.6)
      30/11  10-12  4523 Lab cons
 49    7/12  10-12  4523 Lab cons
       9/12  13-15  L1   Lect  Wrap-up
 50   15/12  14-17  Q22, Q1  EXAMINATION

Reading: Lc Text in lecture notes chapter c. D c.s Demmel text book Chapter c section s.
Contents: What is discussed in this lecture.

Computer lab assignments

The laboratory work consists of three assignments carried out by groups of no more than two students. Each group hands in a report on each assignment at CSC's Students' expedition (NOT the instructors' mail stops!), according to the schedule below. Attach a cover page properly filled in. It can be downloaded here or fetched at the Students' Expedition. We strongly prefer paper hand-ins, the Swedish post office can be relied upon for those unable to attend in person. Homework is handed back in the Student Expedition.

The assignment reports should be clearly written and easy to read for the grader. Provide always an explanation for the observed effects. Mere sheaves of computer plots and printed output of matlab will not be accepted. Here are some guidelines how a report should be prepared. A good example of a nice report can be found here.

Here are the references to the computer labs:

• Lab 1, due November 12, 2010
• Lab 2, due November 26, 2010. Extra tips on sparse matrices in matlab can be found here. The detailed description of the implementation is given in gilbert92sparse.pdf.
• Lab 3, due December 10, 2010. The following files are needed for the assignment: zipdata.mat, ima2.m

See the FAQ concerning the programming assignments!

Recommended Literature

Most of the course is in the book: James W. Demmel, Applied Numerical Linear Algebra, SIAM 1997, Order code OT56, homepage.
   It has been difficult to get the book in Stockholm. There is a way to buy it directly from the publisher, who offers a discount for members. SIAM offers a free student membership to students of SIAM member schools of which one is Royal Institute of Technology, KTH. Sign in at SIAM: Students

Lecture notes

Axel Ruhe: Topics in Linear Algebra. Sold at the students' expedition
Review questions used to prepare for exam will be distributed: RevQ.pdf

Examination

The grading of this part of the course is based on the three programming assignments and a written examn with questions covering the theory part of the course.

We grade the labs up to 6 each and the exam 18, total up to 36. To pass the course you must have submitted all 3 labs and taken the exam with at least 9. If you pass you get a grade according to the accumulated number of points. So grade E corresponds to at least 12 points.