AMSC 660 / CMSC 660 Scientific Computing I, Fall 2010
Dianne P. O'Leary
oleary@cs.umd.edu
Prerequisite:
Undergraduate numerical analysis. Programming assignments
will be in Matlab.
Textbook:
Scientific Computing with Case Studies
by Dianne P. O'Leary,
SIAM Press, 2009.
Topics:
Monte Carlo simulation, numerical linear algebra, nonlinear systems
and continuation method, optimization, ordinary differential equations.
Fundamental techniques in scientific computation with an introduction to
the theory and software for each topic.
Grading:
Based on quizzes, homeworks, and project.
Final Exam:
None.
CMSC Masters Comprehensive Exam grade:
based on quiz grades
Basic Information:
Tentative Schedule for Fall 2010:
01 
Aug 31:  Prelim. 
Sep 2:  Prelim.

02 
Sep 7:  Prelim. 
Sep 9:  Monte C.
HW 1 assigned.

03 
Sep 14:  Monte C. Quiz 1 
Sep. 16:  Monte C. 

04 
Sep 21:  Matrix Comp. 
Sep 23:  Matrix Comp Quiz 2

05 
Sep 28:  Matrix Comp.
HW 1 due. HW 2 assigned. 
Sep 30:  Matrix Comp. 

06 
Oct 5:  Matrix Comp. Quiz 3 
Oct 7:  ODEs 

07 
Oct 12:  ODEs 
Oct 14:  ODEs Quiz 4 

08 
Oct 19:  ODEs HW 2 due.
Term project info available. 
Oct 21:  ODEs

09 
Oct 26:  ODEs Quiz 5 
Oct 28:  ODEs


10 
Nov 2:  Optim. HW 3 assigned. 
Nov 4:  Optim. Quiz 6

11 
Nov 9:  Optim. 
Nov 11:  Optim. 

12 
Nov 16:  Optim. Quiz 7 
Nov 18:  Optim. HW 3 due. 

13 
Nov 23:  Optim. 
Nov 25:  Happy Thanksgiving! 

14 
Nov 30:  Optim. Quiz 8 
Dec 2:  Nonlin. Eqn. 

15 
Dec 7:  Nonlin. Eqn. 
Dec 9:  Nonlin. Eqn. Quiz 9

 Dec 14:  12 noon: Term project due.

2010 Lecture Notes:
Errors and Arithmetic
Dense Matrix Computations
q2.m
Program for a RRQR demonstration.
Changes on p. 23 marked in green.
Optimization
An example of a good linesearch:
cvsrch.m
and
cstep.m
Solving Nonlinear Equations
Homotopy example:
Applying the method to a simple problem, convex optimization
Monte Carlo Methods
Ordinary Differential Equations, Part 1
Ordinary Differential Equations, Part 2
Homework
Homework 1
Homework 2
Homework 3
Quizzes:
For practice, see the sample quizzes below.
Quiz 1 will cover pp. 525 in the textbook, and the Errors and Arithmetic notes.
The first 7 bullets in the "Mastery" list on p. 4 of the textbook give hints.
Quiz 2 will cover pp. 185  219 in the textbook, and the Monte Carlo
notes. The bullets in the "Mastery" list on pp. 185186 of the
textbook give hints. Ignore the section on counting/KRS
and the section on quasirandom numbers. Pay special attention
to Challenge 19.6 and the first 3 parts of Challenge 19.7.
Quiz 3 will cover Sections 5.15.5 of the textbook, and the
dense matrix computations notes through the eigendecomposition.
One question will involve the Gerschgorin theorem.
Quiz 4 will cover the SVD (Section 5.6 of the textbook, pp. 1618
of the notes)
and ODEs (Sections 20.1 through 20.2.4 of the textbook, pp. 116 of
the notes).
Table 20.1 will be given to you if needed, so don't memorize.
Quiz 5 will cover ODEs (Chapter 20, Chapter 21 through Challenge
21.3, and the ODE notes). Pay some attention to the Pointers,
but if you need one of them, I will provide a copy for you to see.
Quiz 6 will cover optimization: Sections 9.1  9.4 and lecture
notes pp. 113.
Quiz 7 will cover optimization: Sections 9.5.1., 9.5.2, 24.2, optimization
lecture notes pp. 1425, nonlinear equations lecture notes pp. 24.
Quiz 8 will cover optimization: Section 9.5.3 and Chapter 10; optimization
lecture notes pp. 2531. One question will involve formulating
a particular image processing problem as an optimization problem,
so familiarity with how an image is stored (see, for example, Chapter 11,
"The Problem")
will be helpful.
Quiz 9 will cover Chapters 2426 and the nonlinear equations notes.
It will be given at 10:20 on Dec 9.
No one may enter the classroom after that time.
Sample quizzes.
Many problems from the old quizzes were
later included as challenges in the textbook.
2010 Term Project Information
CourseEvalUM Fall 2010:
"Your participation in the evaluation of courses through CourseEvalUM is a responsibility you hold as a student member of our academic community. Your feedback is confidential and important to the improvement of teaching and learning at the University as well as to the tenure and promotion process. CourseEvalUM will be open for you to complete your evaluations for fall semester courses in December. Please go directly to www.courseevalum.umd.edu to complete your evaluations. By completing all of your evaluations each semester, you will have the privilege of accessing online, at Testudo, the evaluation reports for the thousands of courses for which 70% or more students submitted their evaluations."