COMP 360: Algorithm Design

McGill University
School of Computer Science
Fall 2022
This Outline Last Updated On: August 17, 2022

General Information
Professor: Robert Robere
Office: McConnell (MC) 309
E-Mail Address: [email protected]
Course Webpage: Administered through myCourses
Lecture Times: Monday and Wednesday 4:05pm–5:25pm.
First Lecture: August 31st, Last Lecture: December 5th
Lecture Room: Strathcona Anatomy and Dentistry (SADB) M-1
Office Hours: TBD
Teaching Assistants: TBD
Prerequisites: COMP 251 or 252, and either MATH 240 or 235 or 363.
Restrictions: Cannot be taken after or concurrently with COMP 362.

Less General Information

Total Number of Lectures: 26
Professor’s Pronouns: He/Him
Professor’s Research Areas: Complexity Theory, Proof Complexity, Lower Bounds
Number of Times The Professor Has Previously Taught Algorithms: 3
Professor’s Favourite Theorem In This Course: Strong Duality of Linear Programming
Last Cool Math Thing The Professor Learned: A one-line probabilistic proof of Erdős-Ko-Rado using Bayes Rule
Album The Professor Is Listening To While Preparing This Outline: Mystic Familiar by Dan Deacon
Favourite Song on the Album The Professor Is Listening To While Preparing This Outline: Sat By A Tree
Professor’s Favourite Movie He Watched In 2022 (So Far): Harakiri by Masaki Kobayashi
Number of Times the Professor Has Previously Watched Kobayashi’s Harakiri: 3

Assessments and Grading Scheme

Type Scheme 1 Scheme 2
Assignments (5) 35% 20%
Quizzes (12) (Weekly, due Fridays, starting September 9th) 15% 10%
Midterm (1) (In-Class, October 19th) 10% 10%
Final Exam 40% 60%

The final grade is the maximum of the two grading schemes posted above. Furthermore, you must
obtain at least a 35% on the final exam in order to pass the course, before curving, in either
grading scheme.

1
Assignments
The assignments in this course will be posted on Crowdmark with their deadlines; the link to Crowd-
mark is reachable through myCourses. In general, assignments will be posted and due on Fridays, with
two weeks for each assignment. Assignments must be submitted on the due date through Crowdmark.
Assignments that are submitted by e-mail will not be accepted. Assignments must be cleanly handwrit-
ten or typeset using a suitable program (e.g. LATEX). If they are handwritten and scanned, please ensure
that the assignment is readable — if a solution is not readable, it will not be graded.
Note that this is a “proofs course”: assignment questions will ask you to formally prove or disprove
various formal mathematical statements. The mathematics background for this course is covered by
its prerequisites; what is essentially required is fluency in discrete mathematics. This means comfort
with things like mathematical induction, graphs, sets, counting arguments, basic discrete probability,
and algorithmic thinking.

Late Policy for Assignments

Late assignments can be submitted after the deadline for up to 3 days, with a 5% penalty for each
late day. (So, since assignments are due by midnight on Friday, they can be submitted until midnight
Monday with a 5% penalty for each day used: submitting Saturday will incur a 5% penalty, submitting
Sunday a 10% penalty, and Monday a 15% penalty). Assignments will not be accepted after the
following Monday evening.
In very special circumstances (e.g. serious illness or medical emergencies for yourself or immediate
family, religious observances, or bereavement), extensions for assignments will be granted; however,
such extensions must be arranged with the instructor. Extensions will not be granted for reasons such
as employment, vacations or other travel, or other assignments due around the due date. For such
reasons you must use the above late policy, and leniancy will not be given. If you find yourself unable
to submit an assignment because of special circumstances, please get in contact with the instructor as
soon as possible.

Quizzes
We will have weekly quizzes administered through myCourses. Each quiz will be released on Monday
and due on Friday, and will require you to answer 1-5 short questions related to that week’s material.
You have unlimited attempts to complete each quiz, and only the maximum score will be taken. 1
Quizzes can not be submitted late.

Textbook
The textbook for this course is Algorithm Design by Kleinberg and Tardos (the version does not mat-
ter). There will be supplementary typed lecture notes that will be posted on myCourses. We will be
following the typed lecture notes more closely than the textbook, but the textbook is extremely good
supplementary material.
1
Don’t sweat this too much, the quizzes are really easy. A quick review of that week’s lecture notes is all that’s needed to
answer the questions.

2
Communication Policy
Communication in this course (outside of lectures or office hours) will be through two channels: the
discussion boards through Ed e-mail. Questions about course content must be posted on Ed, so that
all students of the course can benefit from the answers. Please add descriptive title Please allow
up to 24 hours for your questions to be answered on Ed (in general I will be checking the discussion
board every evening). Any questions about course content sent to the instructor or TAs via e-mail will
not be answered; you will simply be redirected to ask the question on the course forum. Email is the
preferred way to contact the instructor and the teaching assistants for personal or administrative issues.
Please include “COMP 360” in your subject lines, and be descriptive. Finally, please refer to the course
website at least once a week for announcements.

Academic Integrity
McGill University values academic integrity. Therefore all students must understand the meaning and
consequences of cheating, plagiarism and other academic offenses under the Code of Student Conduct
and Disciplinary Procedures (see http://www.mcgill.ca/integrity for more information).
Most importantly, work submitted for this course must represent your own efforts. While the internet
can be an excellent resource for mathematics, you are not allowed to simply search for answers to the
assignment questions; this is cheating. Copying assignments or tests from any source, completely or
partially, allowing others to copy your work, will not be tolerated. Furthermore, you may not upload
any of the course materials to any public-facing website or share the materials outside of the
course.
Having said this: please work with and talk to other students in the course! I absolutely encourage
you to work together — it is the best way to learn. What is important is that the final, submitted work
is your own.

Inclusivity
This course seeks to be a fully inclusive learning environment! That being said, if you experience
barriers to learning in this course, please reach out to me and talk to me about it. This is especially
important given the current remote-learning situation we are dealing with. If applicable, you can also
reach out to the Office for Students with Disabilities (514-398-6009).

Submission of written work in French.

In accord [sic] with McGill University’s Charter of Students’ Rights, students in this course have the
right to submit in English or in French any written work that is to be graded.

Course Outline
Time permitting, we hope to cover the following topics (actual progress may vary slightly from what is
listed below, depending on the semester).

• Network Flows, ∼3 Weeks. Motivation, the Ford-Fulkerson Algorithm and variants. The Max-
Flow/Min-Cut Theorem. Applications, including Bipartite Matching and Scheduling Problems.

3
• Linear Programming, ∼3 Weeks. Motivation and Geometric Intuition. Duality Theory and
Complementary Slackness. The Simplex Algorithm. Applications. Zero-Sum Games and Mini-
max.

• Complexity Theory, ∼3 Weeks. Motivation, Decision Problems and Models of Computation.

Complexity Classes: P, NP, coNP, EXP. Reductions, NP-Hardness and Completeness. The
Cook-Levin Theorem. Applications.

• Randomized and Approximation Algorithms. ∼3 Weeks. Motivation. Randomized Algo-

rithms. Integer Programming and Linear Programming Relaxations. Applications, including
Vertex Cover, Set Cover, Load Balancing, Center Selection. Freivald’s Algorithm and Simple
Randomized Approximations.

I walked off
Sat by a tree
It had captured my attention
When it started asking me:
“What would you cast into existence,
if you contained your persistence to unwind?”
Dan Deacon, Sat By A Tree