CS 559
Instructor: Prof. John Byers
Meetings: MW 9:30 - 11AM
Class Location: PSY B51
Office Location: MCS 270
Office Hours: Mon: 2-3:30, Wed: 11-12:30
Phone: 353-8925
FAX: 353-6457
Email: byers_at_cs_dot_bu_dot_edu
Web: http://www.cs.bu.edu/fac/byers
Scribe Notes
- Lecture 1 covered the Secret Sharing paper (below).
-
Lecture 2, Scribe: Raymond Sweha
-
Lecture 3, Scribe: Konstantin Vodski
-
Lecture 4, Scribe: Sowmya Manjantha
-
Lecture 5, Scribe: Marisa Vasconcelos
-
Lecture 6, Scribe: Praveen Sharma
-
Lecture 7, Scribe: Kevin Donnelly
-
Lecture 8, Scribe: Ilir Capuni
-
Lecture 9, Scribe: Samuel Epstein
-
Lecture 10, Scribe: Michel Machado
-
Lecture 12 PowerPoint slides.
-
Lecture 13, Scribe: Stephen Chao
-
Lecture 14, Scribe: Yin Chen
-
Lecture 15, Scribe: Georgios Smaragdakis
-
Lecture 16, Scribe: Vatche Ishakian
-
Lecture 17, Scribe: Flavio Esposito
Course Overview
Today's complete networking researcher must carry a large toolkit. Expertise
in network measurement, network modeling, protocol design and systems
engineering are perforce. But while many researchers bring these skills
to the table, far fewer have deep insight when it comes to questions of
algorithms and analysis. This is all the more surprising given the wealth
of elegant and essentially algorithmic constructs which have been applied
to networking and Web-related problem domains in recent years. This
graduate seminar will study 1) fundamental algorithmic principles as relate
to networking, 2) how algorithmic methods have been applied to specific
networking applications, and 3) the limits of algorithmic practicability,
i.e. if and when heuristics should be employed.
Prerequisites
The course is geared primarily toward graduate students or advanced
undergraduates who are interested in
pursuing a career in either networking research or algorithms. CS 455/655 or
equivalent is a required prerequisite for this course, as is CS 530
(co-requisite). Masters students and undergrads who did excellent work
in CS 455/655 are encouraged to attend. Most of the papers will
delve deeply into algorithmic, statistical or information-theoretic
techniques, so a graduate level of mathematical sophistication is expected.
Please see the instructor if you are at all uncertain about your level
of preparation.
Course Requirements
For class, students will be expected to read and digest approximately two
research papers per week (prior to lecture). For the majority
of the course, the instructor, along with specialists (see below) will
lead discussions on the current set of papers and will lecture on background
material needed to understand the next set of papers. For each subsection
of the course, a group of students chosen in advance will serve as
specialists, i.e. will be experts on the papers we are discussing,
and will be expected to help facilitate the discussion, brainstorm about research
directions, and help with the presentation of the material (or with supplemental
material).
-
Each student is expected to take scribe notes for one lecture reflecting the
technical material as well as the class discussion. The scribe note style
file will be posted here.
-
A semester-long research project, culminating in a presentation to the
class and a writeup in the style of a conference paper. The project and
presentations will constitute 50% of the overall grade. Suggested
project topics and project deadlines will be announced after the first
few weeks of the course. I will expect students in this class to take
the project very seriously and there will be regular interaction with the
instructor outside of class to work on the projects --- ideally, several/many
of the projects in the class will eventually lead to publishable papers.
-
Class participation will constitute 20% of a student's overall grade
-- this grade will be based both on the student's work as a specialist and
contributions to the class discussion throughout the course.
-
For the remaining 30% of the grade, we will have two in-class quizzes
testing the main concepts in the papers, I will likely pose a few challenging problems
as homework questions, and I will factor in class discussion.