Advanced Topics 2 (236 602):
Parametrized Complexity


Homepage: http://cs.technion.ac.il/~janos/COURSES/FPT/announcement.html

Lecturer: Prof. J.A. Makowsky
Taub 628, Tel: 4358, e-mail: janos@cs

Format: 2 hours lecture

Lecture: Sunday 14:30-16:30
Place: To be announced

Prerequisites: Computability (236 343), Graph Algorithms (234 246)


Course outline:

The course deals witrha methods of classifying and solving problems which are at least NP-hard. The good problems are the Fixed Parameter Tractable Problems (FPT), which are those which can divided into subproblems P(k), which are solvable in polynomial time f(k) n^c with the constant c not dependent on k. There is also a hierarchy of bad parametrized problems, for which it can be shown, that no such algorithms exist unless this hierarchy collapses. We shall develop the theory of parametrized complexity. We shall present a wealth of parametrized problems which are FPT, which includes many variations of SAT, graph problems, and classical optimization problems such as TSP.

Course goal

Leading to the frontier of current research in parametrized complexity.
Introducing topics for M.Sc. and Ph.D. theses.

Course requirements

Projects or take home exam.

Literature

  1. R.G. Downey and M.F Fellows, Parametrized Complexity, Springer, 1999.
  2. Additional literature from Journals.