..
Suche
Hinweise zum Einsatz der Google Suche
Personensuchezur unisono Personensuche
Veranstaltungssuchezur unisono Veranstaltungssuche
Katalog plus

Topics for a Bachelor or Master thesis

Compression of finite structures using automata (topic for master thesis)

Automatic structure are relational structures that are represented in a certain sense by finite automata. Usually, these structures are assumed to be infinite. There is a rich theory of automatic structures relating them to areas like mathematical logic and computability. The concept of automatic structures can be also used in order to represent finite structures in a compressed way by automata. The goal of the project is to compare automatic representations of finite structures with related concepts like BDDs (binary decision diagrams) and to investigate the complexity of algorithmic problems for finite structures when the input structure is given by finite automata.

 

Using SAT solver for algorithmic problems on permutation groups (topic for master thesis)

A permutation on the set {1, 2,…, n} is just a bijective (one-to-one) function on {1, 2,…, n}. The set of all permutations on {1, 2,…, n} form a group, the so-called symmetric group on {1, 2,…, n}, which is also denoted by Sym(n). The multiplication in Sym(n) is just the composition of functions. An important computational problem in this context is the so-called permutation group membership problem:

- The input is: a number n, a permutation f from Sym(n), and a finite subset S of Sym(n).

- The question is whether f can be written as a product (of arbitrary length) of permutations from S.

The classical Schreier-Sims algorithm solves this problem in polynomial time. A more difficult problem is the so-called length problem for permutation groups: The input is the same as in the above permutation group membership problem plus an additional number m. The question is then whether f can be written as a product of permutations from S, where the product must have length at most m. This problem is quite difficult to solve: it is an NP-complete problem (Even and Goldreich 1981) The goal of the project is to explore the use of SAT solver in order to solve instances of the above length problem. SAT solver a software tools that are used for checking whether a given boolean formula is satisfiable (the so-called SAT problem), i.e., whether one can assign truth values to the variables in the formula such that the fomula becomes true. It is not difficult to translate an instance of the length problem for permutation groups into a SAT problem, which then can be attacked with a SAT solver. There are several freely available SAT solver that can be used for this.