46-625 Multivariate Calculus and Linear Algebra

Carnegie Mellon University

August 2003

Instructor: Dr. John Tolle

Email: tolle+@andrew.cmu.edu

Office: Wean Hall 6124

Teaching Assistant:  Albert Cohen; email: aspeight+@andrew.cmu.edu

Questions:  If you have questions about homework, please send email to BOTH me and Albert.  This way, whichever of us receives your message first can respond to it, and this will minimize the time it takes for you to get help.

Fax:  The fax number for the Department of Mathematical Sciences at CMU is 412-268-6380.

Prerequisites: Two semesters of calculus, some background in matrix algebra.

Assignments: There will be four homework assignments. You are welcome (yea, encouraged) to work together on assignments, but each
of you should write up your results separately.  Typically, the problems should be worked out in rough form first; during this stage, you may consort with one another. But a second stage should involve you reviewing and refining your arguments to write up in final form; you should do this part independently. Copying another's work in whole or in part will be considered a serious breach of ethics, but merely discussing problem-solving strategy is OK.

Each assignment will consist of a set of Required Exercises, and a set of (what we call) Elective Exercises, from which you may choose a certain (specified) number to do.  I will read your work and mark weak (or incorrect) points in your arguments, but I will not have time to provide detailed written feedback. Model solutions to at least the Required Exercises will be provided, and because of this, late homework cannot be accepted. On the due date, hand in whatever you've got.

Grading:  To understand the grading in this course, you must understand my philosophy of graduate school:  All graduate students are
A and B students.  There is no excuse for C-level work.  I fully expect that all of you will perform at the A and B levels.  But a graduate
student should be unencumbered by grade anxiety.  It is time for you to put behind your past concern over points, and numbers, and
grades.  It is time to get down to the business of learning, and to demonstrating what you have learned.  In graduate school, this is
what impresses your instructors:  that you can demonstrate and articulate what you have learned.

So you will notice that there is no grading scale.

A final grade of A means that you have demonstrated a satisfactory level of understanding of nearly all of the topics of the course,
with the possible exception of a few topics.  An A grade means that at worst, there are a few minor deficiencies in your understanding
that I believe you could rectify with just a bit more study of those topics.

A final grade of B means that you have demonstrated a satisfactory level of understanding of at least half of the topics of the course
and have demonstrated a clear attempt to learn all of the concepts.  A student who struggles a bit but whose commitment to learning
the material is not in question can fully expect to earn a grade of B.

I've never assigned a final grade of C to a graduate student before contacting the student to discuss the situation and explain the
reasons for the grade.  I don't want to have to do it this time.  A grade of C indicates an unacceptable level of commitment to learn
the concepts, and/or the lack of evidence that the student understands at least half of the material.  Let's make this paragraph
unnecessary; let's have everyone performing at A or B level.

We will return your homework assignments with token scores.  If your score is in the 90s on an assignment, this means that you are
doing well, and you are on track for an A grade.  Keep up the good work.  If your score is in the 80s, this means that there may be
some deficiencies in your understanding that you should try to rectify quickly.  If your score is in the 70s, this means that your level
of understanding (or at least your success in articulating your understanding) is below that of most of your peers.  In this case you
should make an immediate effort to bring your performance to a higher standard.  If you feel that your understanding of the concepts
is intact, then you should try to improve the clarity with which you communicate your work.  Needless to say, then, a score below
70 indicates unsatisfactory performance.

The final grade is determined according to the expectation that roughly half of you will perform at A level and roughly half of you at
B level.  I will look for what appears to be a bit of a gap between the upper echelon students and the lower echelon students.  I will
make an attempt to recognize consistent improvement over the period of the course.

Professor's Appearance: Somewhat haggard.

Exam: There will be one in-class exam, given during the last session, June 27.  We may have one hour of lecture before the exam.
(Of course, you will not be tested on the content of that lecture.)

Schedule

Lecture 1, May 16: Finite and infinite sets, Countability. (Read Chapters 0 and 1.)

Lecture 2, May 23: Limits of functions of one variable, Algebraic structure of R^n, Limits of multivariate and vector-valued functions,
Continuity. (Read Chapters 2 and 3.)

Lecture 3, May 30: I plan to be in NYC on this date. Continuity theorems, Differentiability, The algebra of derivatives, The chain
rule, Mean value theorem, l'Hospital's rule, The Darboux integral. (Read Chapter 4 and Chapter 5 up through Example 5.3.)

Lecture 4, June 6: Integrability theorems, The fundamental theorem of calculus, Convex functions, Jensen's inequality, Analytic functions of one variable. (Read the remainder of Chapter 5, Chapter 6, and Chapter 7 up through Example 7.9.)

Lecture 5, June 13: Partial derivatives, Analytic multivariate functions. (Read the remainder of Chapter 7.) At this point we will enter the
linear algebra portion of the course, and there is no coverage of the topics in the lecture notes. Linear independence, Basis, Dimension,
Orthogonality, Symmetric matrices, Eigenvalues, Eigenvectors.

Lecture 6, June 20: Eigenbases, Quadratic forms, Positive definiteness.

Lecture 7, June 27: Possibly, and probably, one hour of lecture before the exam.