This page is under construction.

CMSC 36500 = Math 37500

Algorithms in Finite Groups:
The Graph Isomorphism problem

Spring 2017

TuTh 12:00-1:20, Ryerson 276

Instructor: László Babai

NEW: Reading

Instructor's paper Graph Isomorphism in Quasipolynomial Time, Version 2.1 (May 23, 2017). This version represents a significant but incomplete update of Version 2, posted on arXiv. (See WARNING on the paper's front page and in the Acknowledgments.) The paper is referenced as [GIQ] in the Class 17 notes.

Grading

Graded homework 40%
Quiz (May 16, 25 min) 12.5%      see statistics
Test (May 30, 75 min) 42.5%
Class participation 5%

A REQUEST

Please send email to the instructor with answers to the following questions. Please answer the questions even if you do not require a grade but you are attending the class.

Please respond by Friday, May 5.

Class notes by Robert Green and Angela Wu, including homework (not verified by instructor unless expressly stated otherwise)

Class 18 - May 25 due May 30. Verified notes.

Class 17 - May 23 due May 25. Detailed, verified notes.

Class 16 - May 18 due May 23. Detailed, verified notes.

Class 15 - May 16 due May 18. Detailed, verified notes.

Class 14 - May 11 due May 16. Detailed, verified notes.

Class 13 - May 9 due May 11. Partially verified notes.

Class 12 - May 4 due May 9. Detailed, verified notes.

Class 11 - May 2 due May 4. Detailed, verified notes.

Class 8 - April 20 due April 25. Detailed, verified notes.

Class 7 - April 18 due April 20

Class 6 - April 13 due April 18

Class 5 - April 11 due April 13

Class 4 - April 6 due April 11

Class 3 - April 4 due April 6

Class 2 - March 30 due April 4

Class 1 - March 28 due March 30

The course may be of interest to those interested in finite groups, combinatorics, and the interaction of these fields with the theory of algorithms.

Synopsis

The main goal of the course is a full presentation of the quasipolynomial-time algorithm for the Graph Isomorphism problem I announced in November 2015.

Toward this goal, we shall spend much of the time studying

This is a pure mathematics course, no prior knowledge of the theory of algorithms is required.

The prerequisite is basic familiarity with finite groups, on the level of the group theory segment of an undergraduate abstract algebra course: be comfortable with the proofs of Sylow's theorems and the Jordan--Hölder Theorem.

The course will begin with an introduction to elements of group theory beyond the normal undergraduate curriculum, including solvability, nilpotence, and classes of finite simple groups.

The material will overlap with my 2014 course by the same course number. You may want to check out the exercises posted there.