Math 147

Linear Algebra for Applications

Class meets daily, 11:30 AM to 1:00 PM in Phillips 367, Summer Session I, 2005
Instructor: Sorin Mitran
Office hours: Mo-Th, 10:30 AM to 11:20 AM, Fr, 1:00-1:30 PM, Phillips 307

[Motivation] [Syllabus] [Grading] [Text] [Homework] [Computer work] [Final]

Motivation and objectives

Linear algebra arises in surprisingly many fields of the sciences and within mathematics itself. Want to find the effectiveness of a drug against a population sample? A linear system will typically arise. Determined to optimize the flight path of an airliner - again linear problems present themselves. The purpose of this class is to present the basic mathematical theory from linear algebra and study how these concepts can be productively used.

Syllabus

  • Linear systems - existence of solutions, Gauss elimination, Gauss-Jordan algorithm, pivoting, rectangular systems, echelon forms

  • Matrix algebra - matrix addition, multiplication, inverse, block multiplication, LU factorization

  • Vector spaces - linear independence, basis and dimension, rank, linear transformations, norms, scalar product, orthogonality, Gram-Schmidt procedure, QR factorization

  • Determinants - definition, elementary properties, Cramer's rule

  • Eigenvalues and eigenfunctions - elementary properties, diagonalization, ODE systems

Grading Policy

There are 24 lessons in the course. Grading is based upon homework and the final examination. Eight homeworks shall be issued, one every three lessons. Each homework is worth 10 grade points. The final examination is worth 20 grade points. Two extra credit assignments worth 8 points each shall be given after lesson 12. In total 116 points are possible. Grade points are translated to letter grades according to the following table:

B+

89-92

C+

75-80

D+

57-62

A

97-100

B

85-88

C

69-74

D

50-56

A-

93-96

B-

81-84

C-

63-68

F

0-49

but a minimum final examination score must also be obtained. There is no "grading on a curve" for this course.

Minimum examination grades

The final examination serves as a verification of the student's mastery of basic concepts and procedures. A student must obtain at least 10 points on the final examination; otherwise a grade of F is entered in the student's record. The examination shall contain 12 questions worth 2 points each from which the student may select 10 to answer. Of those selected, 50% must be answered correctly in order to obtain a passing grade. The examination is structured so that 50% of the questions shall represent basic concepts while the balance shall treat more advanced techniques.

Course Text

The course text is:

Matrix Analysis and Applied Linear Algebra, by Carl D. Meyer, SIAM, 2000 ISBN 0-8987 I-454-0

Homework

Homework is to be turned in at the end of the class held on the homework due date. Late homework is not allowed, examined or graded. Please be aware that a summer course has a rapid pace to it. Not applying the required effort for even a few days will lead to difficulties in obtaining a good grade. It is essential for a student to start the homework assignment on the day it is issued.

HW #

Given

Due

Assignment

Solution

1

5/17

5/20

Meyer: 1.2.1 - 1.2.10

HW1.pdf

2

5/20

5/25

Meyer: 1.3.1-1.3.3, 1.4.1, 1.4.3, 1.5.1:(a) - (e)

HW2.pdf

3

5/25

5/31

Meyer: 1.5.4:(a)-(d), 1.6.1: (a)-(f)

HW3.pdf
fpGauss.m
fp.m
scalerows.m
pivotrows.m

4

5/31

6/2

Meyer: 2.1.1:(a)-(b), 2.2.1:(a)-(b), 2.3.1:-(a)-(b), 2.4.1:(a)-(b), 2.5.1:(a)-(b)

HW4.pdf

5

6/2

6/7

Meyer: 3.2.3:(a)-(c), 3.2.6:(a)-(b), 3.3.1:(a), (c), (e), 3.4.2, 3.4.3

HW5.pdf

6

6/7

6/10

Write a Matlab code to solve linear systems through LU factorization. (1 point for the LU factorization, 1 point for forward substitution, 1 point for back substitution, 1 point for testing the code on systems with Hilbert matrices of sizes 5,10,...,50). Meyer: 3.5.4 (1 point for (a)-(c), 3.5.8. (a) & (b) (1 point), 3.5.8. (c), 3.6.3, 3.7.10 (a)-(b)

7

6/10

6/15

Meyer: 4.1.7, 4.1.8, 4.1.12, 4.2.6, 4.2.9, 4.2.12, 4.2.13, 4.3.5, 4.3.6, 4.3.12

8

6/15

6/17

Meyer: 4.4.8, 5.1.5, 5.3.4, 5.4.1: (b), (d), 5.4.8, 5.4.10: (a), (b), 5.5.1: (a), (b)

Extra credit projects

Project 1: Read Example 3.2.1 from Meyer. Simple ensembles of L springs and M point masses are often used as models in chemistry, engineering, physics. Write a Matlab code that computes the stiffness matrix and its inverse (Gauss-Jordan algorithm) after reading the known equilibrium positions and stiffnesses for L three-dimensional springs (2 points). Assume reasonable connectivity between the springs, e.g. consider them representative of molecular bonds. Consider that the ensemble is changed by adding one more spring element within the network. Use the Sherman-Morrison formula to get an update of the inverse of the stiffness matrix (4 points). Now consider a sequence of such updates. Write the Matlab code that reads in a new spring position and incrementally updates the stiffness matrix for L' additional springs.

Project 2: Read about rotations in 3D space (Meyer, p. 328). Consider now a reduced Rubik cube formed of 8 subcubes, i.e. instead of the variety you pick up at a toy store which has 27 = 3 x 3 x 3 subcubes, the one considered here is 2 x 2 x 2. Given some initial arbitrary orientation of the colored faces of the subcubes, write a Matlab code that uses reflection operations to solve the Rubik puzzle, i.e. have all faces of the large cube of the same color.

Extra credit projects are due on June 20.

Computer work

Practical computing work is an essential part of applied linear algebra. We shall use Matlab to carry out programming exercises needed to study linear algebra algorithms. An introduction to Matlab capabilities will be given on 5/18, 5/24 & 5/25 largely following the Matlab primer.

Final examination

The Final Examination will be given in Phillips 367 on Monday, June 20 from 11:30 AM to 2:30 PM.

Typical final examination questions

Basic concepts

Meyer: 1.2.10, 1.2.16, 2.1.1, 2.3.3, 2.4.1, 2.4.2, 2.5.4, 3.5.4, 3.6.8, 3.7.5, 3.7.8, 3.9.6, 3.10.2, 4.1.2, 4.1.7, 4.1.10, 4.2.5, 4.3.6, 4.4.4, 4.7.1, 4.7.14, 4.8.7, 4.8.8, 5.2.6, 5.4.3, 5.4.6, 5.5.2, 5.5.3, 5.5.5, 5.6.2, 5.6.4, 5.6.5, 7.1.1, 7.1.2, 7.2.1

Advanced techniques

Meyer: 3.7.11, 3.9.9, 3.10.3, 4.1.12, 4.2.7, 4.2.12, 4.3.8, 4.3.12, 4.3.13, 4.4.10, 4.4.16, 4.5.9, 4.5.12, 4.7.17, 4.8.10, 5.2.7, 5.4.9, 5.4.15, 5.5.11, 5.6.11, 7.1.8, 7.2.2