Mathematics E-23a, Fall 2021

Linear Algebra and Real Analysis I

Last revised: 2021–11–19
Class Meetings: Online live web conferences to be held
• Thursday 17:30–19:30
• Friday 19:30–21:30
• Saturday 09:00–11:00
• Sunday 15:00–17:00
All times are Eastern US. You are expected to choose one time and attend every week.
Instructor: Grant Murray (to be addressed as “Grant”, please)
Email: cael@alum.mit.edu
Instructor Office Hours: Saturday 11:30–12:30, Saturday 19:30–20:30
Teaching Assistants:
Head TA: Michael Liotti, michaeljliotti@gmail.com
Éminence Grise: Kris Lokere, krislokere@gmail.com
Consigliere: Lara Zeng, lzeng@college.harvard.edu
Deep Learner: Benjamin Basseri, basseri@cs50.harvard.edu
Wizard: Constantine Tsibouris, ctsibouris@college.harvard.edu
TA Office Hours:
Michael: Monday, 7–9 p.m.
Kris: Friday, 7:30 – 9 a.m. (yes that means morning in Eastern US time)
Lara: Sunday, 2–3 p.m. and 5–6 p.m. (Same as Sunday meeting link)
Ben: Monday, 4–6 p.m.
Constantine: Friday 2–3 p.m, Monday 1:30–2:30 p.m
Course Website: https://canvas.harvard.edu/courses/94697

Summary
Goals. Math E-23a is a clone of Math 23a, the first half of a moderately rigorous Harvard
College course in linear algebra and vector calculus. It is designed for students who are
serious about mathematics and interested in being able to prove the theorems that they
use, but who are as much concerned about the application of mathematics in fields like
physics and economics as about “pure mathematics” for its own sake. Trying to cover
both theory and practice makes for a challenging course with a lot of material, but it is
appropriate for the audience.

Target Audiences. There are two major audiences. The first group are students looking
to strengthen their background in proof-based mathematics in preparation for graduate
work in fields like machine learning, economics, applied mathematics, or data science.
The second, smaller group are secondary-school students who are moving beyond Ad-
vanced Placement calculus, perhaps to convince the colleges to which they are applying
that they can handle a rigorous proof-based course.
1
Prerequisites. This course is designed for the student who received a grade of 5 on the
Calculus BC Advanced Placement examination or an A in a college course in single-variable
calculus like Math E-16. The most important prerequisite is the attitude that mathematics
will be more fun and more exciting when it is no longer a plug-and-chug activity.
Our assumption is that the typical Math E-23a student knows (or remembers) only
high-school algebra and single-variable calculus and is currently better at formula-crunching
than at doing proofs. We do not assume that Math E-23 students have any prior experience
in linear algebra beyond solving systems of linear equations in high school algebra, and we
assume no knowledge of multivariable calculus.

Course Structure (in Brief)

Subject Matter: The course is divided into three Dynasties of roughly equal length.
Linear Algebra Dynasty: Covering chapters 0, 1, and 2 in Hubbard & Hubbard.
Real Analysis Dynasty: Covering chapters 1, 2, 3, and 5 in Ross.
Vector Calculus Dynasty: Covering chapters 2 and 3 in Hubbard & Hubbard.

Lectures: In Fall 2015 all 23a lectures were captured on video. Watch the lecture videos
before the weekly class meeting.

Lecture Outlines: Lecture outlines have blank spaces that you can fill in as you watch the
lecture videos. There are also edited copies of the notes from which Paul lectured.
Different students learn in different ways, but most students benefit from taking
their own notes on the blank spaces while watching the recorded lectures.

Textbooks: There are two required textbooks.

• John H. Hubbard and Barbara Burke Hubbard, Vector Calculus, Linear Algebra,
and Differential Forms: A Unified Approach, Fifth edition, Matrix Editions.
http://matrixeditions.com/5thUnifiedApproach.html
In E-23a we will cover only Chapters 0–3. If you plan to continue in Math
E-23b or E-23c in the spring, be sure to get the current (fifth) edition. A used
copy of the fourth edition will suffice for the fall term.

• Kenneth A. Ross, Elementary Analysis: The Theory of Calculus, 2nd ed., 2013.
This will be our primary text for single-variable real analysis. It is available
electronically through the Harvard library system.

Course Meetings: Online course meetings will be weekly for two hours. The first hour
(“Presentations”) will be student presentations of key definitions and theorems,
while the second hour (“Problems”) will be breakout group problem solving.

Homework: ∼Eight questions weekly, due on Tuesdays by 23:59 Eastern US time.

Proofs: Most lectures will include one or more proofs that you are expected to learn well
enough so that you can do them without notes.
2
Exams: You will need an informal proctor for the midterm exams. The proctoring policy
for the final exam is undetermined as of this writing.

Exam When What

Exam One Early October Linear algebra
Exam Two Early November Real analysis
Final Exam Mid-December Vector analysis and calculus

R: Required only for students who enroll for graduate credit, optional for everyone else.

Technology for online learning: We will use Zoom. You need a computer with a
microphone and webcam.
You need at least occasional access to a printer. You will need to print the
exams so that you can take them with your computer switched off.
You need to be able to convert your homework and exams to PDF files. A scanner
is ideal, but a smart phone with a good camera will suffice.
You will do mathematical presentations in Zoom. Using a tablet works well, or
you can use your webcam and write on a whiteboard. There is a link on the course
website to a video in which students demonstrate several styles for mathematical
presentations on Zoom. For group problem-solving, at least one member of each
breakout group will need to be able to display and upload handwritten solutions.

YouTube videos: The Lecture Preview Videos were made by Kate Penner. They cover
the Executive Summaries in the weekly course materials, which go over all the
topics, but without proofs or detailed examples. If you watch these videos (it takes
about an hour per week) you will be very well prepared for the lecture videos.
The R script videos were made by Paul. They provide a line-by-line explanation
of the R scripts that accompany each week’s materials.

Me: I have a bachelor’s degree in mathematics from MIT and a master’s degree in economics
from BU. I have been a teaching assistant for Math E-23a, E-23c, E-158, E-210,
and E-216, and course head for Math E-216 in 2019 and 2021. I will be course head
for Math E-23b and Math E-216 in spring 2022.

Grades: Assuming that you opt in to section and proof logging, your course grade will be
determined as follows:
• Homework, 50 points. Your worst (most points lost) score will be converted to
a perfect score.
• Presenting and listening to proofs, worth up to 35.7 points but only 32.3 goes
into the denominator
• Two midterm exams, 50 points each
• Final exam, slightly more than 60 points
• Class participation, 20 points (volunteering for presentations, participating in
group problem-solving)
• R bonus points, required for graduate credit students, optional for everyone
else. Added nonlinearly — you will only benefit from doing R. R is worth
3
 roughly 62 points in the numerator, and the greater of {95% of your R points,
50% of the available points} in the denominator. For undergraduate credit
students we calculate your grade twice; once with the R and once without, and
take the higher. The R points come from submitting R Workshop problems
(8 points), submitting R homework problems (∼ 30 points), and doing two R
take-home tests (∼ 12 points each).
If you opt out of section and/or proof logging, those elements will be dropped
from both numerator and denominator.
The grading scheme:
Points Minimum Grade
94.0% A
89.0% A-
84.0% B+
78.0% B
72.0% B-
65.0% C+
58.0% C
51.0% C-
There is no “curve” in this course! You cannot do worse because your classmates
do better. If every student demonstrates mastery of the material in the class, it is
possible for everyone to receive an A.

4
 Week-by-week schedule

Dynasty Date Topic

1 2–13 September
1 13–19 September
1 20–26 September
1 27 September–3 October
2 4–10 October
9–10 October
2 11–17 October
2 18–24 October
2 25–31 October
3 1–7 November
6–7 November
3 8–14 November
3 15–21 November
25 November
3 23 November–2 December
13–18 December
Fields, vectors and matrices
Dot and cross products; Euclidean geometry of Rn
Row reduction, independence, basis
Eigenvectors and eigenvalues
Number systems and sequences
Exam One on Linear Algebra Dynasty
Series, convergence tests, power series
Limits and continuity of functions
Derivatives, inverse functions, Taylor series
Topology, sequences in Rn , linear differential equations
Exam Two on Real Analysis Dynasty
Limits & continuity in Rn ; partial & directional derivatives
Differentiability, Newton’s method, inverse functions
Thanksgiving
Manifolds, critical points, Lagrange multipliers
Final Exam on Vector Calculus Dynasty

Official Harvard Extension School Policies

Disability and Accessibility. The Extension School is committed to providing an accessible
academic community. The Accessibility Office offers a variety of accommodations and services
to students with documented disabilities. Please visit https://www.extension.harvard.edu/
resources-policies/resources/disability-services-accessibility for more information.

Academic Integrity.
You are responsible for understanding Harvard Extension School policies
on academic integrity and how to use sources responsibly. Not knowing the
rules, misunderstanding the rules, running out of time, submitting the wrong
draft, or being overwhelmed with multiple demands are not acceptable excuses.
There are no excuses for failure to uphold academic integrity. To support your
learning about academic citation rules, please visit Harvard Extension School
Tips to Avoid Plagiarism, where you’ll find links to the Harvard Guide to
Using Sources and two free online 15-minute tutorials to test your knowledge
of academic citation policy. The tutorials are anonymous open-learning tools.

Course Policies
Homework Collaboration & Citation Policies. You are encouraged to discuss the
course with other students and the course staff, but you must always write your homework
solutions out yourself in your own words. If you collaborate with classmates to solve
problems that call for R scripts, create your own file after your study group has figured out
how to do it.
You must list all your collaborators on each homework assignment.
5
 Proofs that you submit to the course website must be done without consulting files that
other students have posted.
If you have the opportunity to see a complete solution to an assigned problem, please
refrain from doing so. If you cannot resist the temptation, you must cite the source, even
if all that you do is check that your own answer is correct. You are forbidden to upload
solutions to homework problems, whether your own or ones that are posted on the course
website, to any publicly available location on the Internet. Multiple students have been
subject to harsh academic discipline for violating these policies in the past.
Anything that you learn from lecture, from the textbook, or from working homework
problems can be regarded as “common knowledge” for purposes of this course, and the
source need not be cited. Anything learned in prerequisite courses falls into the same
category. The intended effect of this policy is that you can avoid wasting time trying to
cite sources on exams, but you will still be complying with the requirement of “accurate
attribution of sources” since common knowledge does not have to be cited.

Course Structure (in Detail)

Lectures. Even though Paul Bamberg is not directly involved in teaching the course this
year, you may still end up learning most of the material from him. This is because the core
of our material content are his recorded lectures from Fall 2015. Recorded lectures give
you the ability to make extensive use of speed controls and rewind, features sadly lacking
in real life.
You are strongly encouraged to print the lecture outlines, and use the blank spaces to
take notes by hand while viewing the lectures. Paul has been teaching for something like
fifty years, so he can be pretty quick at the blackboard. Fortunately, you can slow the
video playback to three-quarter or even half speed, and rewind if you miss something.

Textbooks. The textbook by John and Barbara Hubbard is intended as a treatment of

linear algebra and vector calculus as a single subject. I wish I had been taught in this
fashion rather than the more traditional approach that treats linear algebra and vector
calculus as essentially unrelated and relegates them to entirely separate courses.
The Hubbards are especially interested in the relationship between proofs and algorithms,
and you may find that their proofs read something like computer programs and their
computer algorithms read like mathematical proofs. Those students with plenty of coding
experience but little exposure to proofs can benefit from understanding that a rigorous
approach to mathematics is not an entirely unfamiliar way of thinking.
The Hubbards describe the relationship between calculus and analysis by saying that
calculus is like learning to drive a car, while analysis is like learning to design and build a
car. Our principal resource for real analysis, the book by Ross, was originally introduced
to the course when Paul Googled “easiest real analysis textbook”. It has been referred to
as a “gentle,” “hand-holding” text. Nonetheless, even a gentle, hand-holding introduction
to designing and building a car should not be taken lightly.

Classes. Classes will meet via Zoom. There are four sections, one each on Thursday,
Friday, Saturday, and Sunday at the end of the week in which you watch the lectures. If
one or more of the times are especially popular or unpopular, we may adjust. You will be
assigned a section based on your preferences. Please try to attend the same section every
6
week if possible. Occasional visiting-scholar arrangements can be made, if you are getting
married or singing the lead in a major production of Shostakovich’s The Nose.
If you prefer, you can opt out of class participation. In that case you do not attend
class, and homework and exams will count for a greater fraction of your grade. You may
of course attend office hours to get help with the homework, and you will have access to
the solutions to problems that were solved in class. In my opinion, if you opt out of class
participation you are giving up a valuable experience while still paying the same tuition as
those students who do participate, but the option is available.
The teaching assistants will hold online office hours, at which students can get together
to discuss homework problems and present proofs to one another. Participation is optional
but is strongly recommended.
Presentations. In the first hour, students will present, seminar-style, five or six pivotal
topics from the lectures for that week, which go over key definitions and proofs.
The student presentation format has been used in most of Paul Bamberg’s classes in
the last few years, with mixed success. If students are diligent about their preparation,
developing a deep enough understanding of their topic that they can present the material
in their own idiom, their explanations are often as lucid as Paul’s and can substantially
clear up confusion for their classmates. If students are less diligent, parroting the section
of lecture that they are presenting on without really making the material their own, they
do neither themselves nor their classmates much good at all.
Trying to squeeze six presentations into an hour is challenging, and requires students to
polish their delivery; please actually go through the presentation the night before, timing
yourself, to make sure you can get the key points across in at most eight minutes, which
allows time for your classmates to interrupt you and ask questions. The skills necessary
don’t have much to do with math, but they are invaluable in any position in any field.
Each week you will volunteer to present a topic, using a shared spreadsheet. You can
volunteer for a topic that someone else has already signed up for, but if there are topics no
one has volunteered for, you cannot be the third to choose a popular topic. When class
meets, as each topic arises one student will be randomly selected from among the volunteers
to do the presentation.
Your fellow students will enthusiastically interrupt you with questions, and you should
do the same for them. Everyone has already sat through a lecture on this material, so don’t
waste the opportunity to make sure you really understand.
Extension students lead busy lives and often need to travel. So you only have to volunteer
to present a seminar topic in 10 of the 13 weeks. If you are out of town two or three times
and send the section instructor a one-line email explaining the circumstances, you can still
get full credit for participation.

7
Problems.
Mathematics is not a spectator sport.
George Pólya

You need to actively engage with the material. During the second hour of class meetings
students will be broken into groups of two to four and given problems to work on together.
Some problems are nothing more than examples from lecture but with different numbers,
while others are more challenging.
Group problem solving if done well can be an astonishingly efficient use of class time,
but if done poorly can be almost worthless. Just as with the presentations, this depends
directly on the effort the students put in. In the one hour given, the group is expected to
attempt at least two problems and write up their work to post it on the course website.
If you don’t solve a problem, that’s OK, but you do need to upload a write-up of your
attempt.
A note — with both seminar presentations and group problems, the ability to write
math in a way that can be shared in real time online is crucial. In the past, students have
employed all kinds of solutions to this problem. Some have just held a phone camera in the
air with one hand, taking video of the other hand writing on a piece of paper. Others have
made use of various more technologically sophisticated approaches. Your fellow students are
your best resource here; some of them are aware of much better technological solutions than
am I. There is a video on the Canvas website showcasing some of the different presentation
styles that have been used in the past, but it would be impossible to demonstrate the full
range of approaches.

Homework. The assignment is included in the same online document as the lecture notes
and group problems. The homework is an integral part of the course and does count for a
meaningful portion of your grade. The problems are in general harder than the workshop
problems, and certainly harder than the exam problems.
Many assignments will include a couple of optional problems whose solutions require R
scripts. These scripts should be uploaded electronically to the website each week. Please
include your name as a comment in the script and also in the file name.
Your homework should be uploaded as a single .pdf file to the course website. It can be
typeset or it can be scanned or photographed and converted to .pdf. It should be legible,
as judged by me. For both your professional development and my own eyesight, it would
be good to become adept with some flavor of TEX: LATEX, pdfTEX, XeTEX, LuaTEX, etc.
Anyone intending to continue in academia will eventually need to be conversant with some
variety of TEX.
If you do submit handwritten work, it should be a final, corrected version with minimal
scratch-outs. The extra time required is small compared with the overall time it will take
to do the problems. Especially for proofs, presenting your work in an organized manner is
a necessary part of the process. More generally, when writing or speaking mathematics,
your presentation should be driven by the needs of your audience. If your audience is your
fellow students, your goal is to explain your ideas such that they can be grasped easily. If
your audience is your instructor or teaching assistants, your goal is to convince us that
you know what you’re doing. If you’re writing a paper for publication in a peer-reviewed
8
journal, your goal is to make your theorem look very important. The styles you use should
reflect your goals.
Homework that is handed in after the day when it is due will not be graded. If it looks
fairly complete, you will get a grade of 50% for it.
Of course, most of us lead fairly complicated lives, and our external commitments can
interfere with submitting assigned work on time. I would like to be able to be reasonable
about allowing for the vagaries of our lives. However, I am a very gullible person and
liable to believe any excuse someone might invent. As a result I am not well suited for
the business of judging the validity of explanations for late work. Hence this policy: your
lowest homework score will be converted to a perfect score before calculating grades. Keep
this policy in your back pocket for emergencies, rather than using it just because you are
somewhat busier than usual.
Students may appeal homework grades; an appeal will be regraded from scratch. The
regraded score could rise, fall, or remain the same.

Proofs.
Logic is the hygiene that the mathematician
practices to keep [their] ideas healthy and strong.
Hermann Weyl

If your math background has been focused on learning techniques for solving problems
(which tends to be the most common approach for courses aimed at scientists and engineers),
you have missed out on one of the most beautiful aspects of mathematics. Proofs save us
when our intuition is insufficient. A good proof convinces and enlightens. A very good
proof convinces, enlightens, and highlights the necessary assumptions, which lets us ask
the next question: what happens if these assumptions don’t hold?
Each week’s course materials includes two proofs. Often these proofs appear in the
textbook and will also be covered in lecture. You can earn points towards your grade by
presenting these proofs to teaching staff and to each other without the aid of your course
notes. Here is how the system works:
When we first learn a proof in class, only members of the teaching staff are “qualified
listeners.” Anyone who presents a satisfactory proof to a qualified listener also becomes
qualified and may listen to proofs by other students. This process of presenting proofs to
qualified listeners occurs separately for every proof.
You are expected to present each proof before the date of the quiz on which it might
appear; so each proof has a deadline date. Distance students may reference the additional
document which details how to go about remotely presenting proofs to classmates and
teaching staff.
Here is the grading system:
• Presenting a proof to any qualified listener: 0.95 points before the deadline, 0.8
points after the deadline. You may only present each proof once.
• Listening to a fellow student’s proof: 0.1 point. Only one student can receive credit
for listening to a proof presentation.
9
Students who do the proofs early and listen to lots of other students’ proofs can get more
than 100%, but there is a cap of 35.7 points total. You can reach this cap by doing each
proof before the deadline and listening twice to each proof.
Either you do a proof right and get full credit, or you give up and try again later. There
is no partial credit. It is OK for the listener to give a couple of small hints.
You may consult the official list of proofs that has the statement of each theorem to
be proved, but you may not use notes. That will also be the case when proofs appear on
quizzes and on the final exam.
It is your responsibility to use the proof logging software on the course website
to keep a record of proofs that you present or listen to.
Each midterm exam will include two proofs chosen at random from a list of 8–16 proofs
that will be distributed in advance, and three proofs will appear on the final exam.
Students occasionally ask if they are being asked to “memorize” proofs. If you have the
ability to memorize an entire page of text in a language you don’t speak, then your mind
works differently from most. It is much easier to remember these proofs by understanding
them than by memorizing an arbitrary sequence of equations and words.
Exams. There will be two midterm exams, one in early October and a second in early
November. For each, you will need to find an informal proctor (a roommate or family
member is fine) who will certify that you did the exam under closed-book conditions, with
computers and cell phones turned off. You must complete the exam in a single sitting during
the 24-hour exam window, but otherwise there is no time limit. Students usually finish in
about two hours, but there is wide variance around that expectation. The completed exam
should be scanned and uploaded in the same manner as homework.
Midterm exams will include questions that resemble the ones done in class, and each will
include two randomly-chosen proofs from among the numbered proofs in the relevant weeks.
There may be other short proofs similar to ones that were done in lecture and problems
that are similar to homework problems.
There will be a regular three-hour non-cumulative (i.e., covering material from only the
final third of the course) final exam in December. The final exam will have three proofs.
The proctoring policy for the final exam is undetermined as of this writing.
Alternatively, you may opt out of the proof logging system. In that case you will need to
do one more proof on each midterm exam.

10
Useful software.
• R and RStudio
This is required only for Extension students who register for graduate credit, but it
is an option for everyone. Consider learning R if you
– are interested in computer science and want practice in using software to do
things that are more mathematical than can be dealt with in introductory CS
courses.
– plan to take Math E-23c in the spring term.
– are thinking of taking a statistics course, which is likely to use R.
– are hoping to get an interesting summer job or summer internship that uses
mathematics or deals with lots of data.
– want to be able to work with large data files in research projects in any field
(life sciences, economics and finance, government, etc.)
R is free, open-source software. Instructions for download and installation are on
the website. You will have the chance to use R at the first class, so install it right
away.
On the course Website are a set of R scripts, with accompanying YouTube
videos, that explain how to do almost every topic in the course by using R. These
scripts are optional for undergraduate-credit students, but they will enhance your
understanding both of mathematics and of R.
Use of R. You can earn “R bonus points” in three ways:
• Participating in an R workshop in which students solve problems that require
creation of R scripts. These will be available for classes 1–4 and 9–12. (about 8
points)
• Submitting R scripts that solve the optional R homework problems (available for
most units). (about 30 points)
• Doing two R review tests, one for linear algebra, one for multivariable calculus.
(∼ 24 points)
If you earn more than half of the possible R bonus points, you will benefit from the
“graduate credit” grade calculation. We will add in your R bonus points to the numerator
of your overall course score. To the denominator, we will add in 95% of your bonus points
or 50% of the possible bonus points, whichever is greater.

11