This course on advanced algorithms will cover approximation algorithms and randomized algorithms and perhaps other topics to be determined. The structure will be much like CS515, Fall 2010, with emphasis on problem solving. We will be using the textbook Approximation Algorithms by Vazirani (corrected 2nd printing, 2003), Tim Roughgarden's lecture notes and Luca Trevisan's lecture notes. Other (free) materials may be provided throughout the course.

My office hours are 10-11 AM Tuesday and Thursday. For other times, please email to schedule an appointment.


This schedule is tentative and subject to change (particularly in later weeks) based on the pace of the class and the difficulty of the problem solving sessions.

Week Tuesday Thursday Other
Ch. 1 and 2 (introduction)
min-read: sec 1.0-1.1, 2.0-2.2
introduction continued Hardness of approximation of vertex cover (FYI only):
No 1.3603-approx for vertex cover unless P=NP.
No ~2-approx for vertex cover unless UCG is false.
H(n) < ln(n) +1
Problem solving session A
Questions (1) 1.7, (2) 2.11, (3) 2.13
LP (duality)
read: Chapter 12
prereq: max-flow, min-cut
linear programming
Written solutions due. See here? for details.
Approximation via LP rounding
read: Chapter 14
Problem solving session B Struggling with written English?
Take your write-ups to the writing center.
Problem solvig session B, cont. Ch. 13 (dual fitting) Monday: 4PM seminar by Gordon Wilfong
5 Ch. 15 (primal-dual)
Session B written solutions due.
Problem solving session C
6 Multi-way cut, Ch. 4 and 19
Gomory-Hu cut tree?
Multi-way cut, cont.
Detailed analysis of multiway cut (Renato Werneck)
7 Problem solving session D TSP, Steiner tree, Ch. 3
8 Steiner forest, Ch. 20
Another write-up of Steiner Forest (Laura Poplawski)
Session D written solutions due.
Problem solving session E
9 Approximation schemes and planar graphs
Dynamic programming review (Jeff Erickson)
Baker's technique: Designing approximation schemes for planar graphs
Corrections/comments on the above notes are welcome
Session E written solutions due.
10 Problem solving session F
Final exam/project/assignment will be released.
class cancelled.
11 Final project due Monday at 5PM, emailed to me as a pdf.


Your grade will be determined as follows:

Problem solving sessions (oral presentation and participation) 25%
Problem solving sessions (written solutions) 25%
Final project 50%

Please refer to the collaboration and academic honesty? and grading and late policies?.

Your grades indexed by code are available [ here]. Your code is generated by multiplying the middle three digits of your OSU ID number by the last three digits of your OSU ID number.

Problem solving sessions

Prior to the problem solving session, a set of problems will be posted. You will be (randomly) assigned a subset of these problems for which you are responsible for solving. Please work with each other in solving these problems! It is highly encouraged that you attempt to solve all of the problems, or at least consider how you would start to solve all of the problems.

During the problem solving session, we will discuss the solutions to the problems until we are sure of a completely correct solution. Due to time constraints, we may not complete all the problems. Participation is highly important in this step. Whether or not you were responsible for a problem, you should be asking questions, suggesting improvements or clarifications. One quarter of your grade will be based on this participation. (I leave it up to you to make sure I know your name!)

After the problem solving session, you will be (randomly) assigned a problem to prepare a (complete, formal, concise) written solution, due by the start of the following class meeting. This problem will be different from one of the problems you were responsible for solving (hence the suggestion to attempt all problems), so you should pay good attention during the problem solving session! You may not be asked to write a solution after each problem solving session, as I hope to give very thorough feedback.