Detailseite
Projekt Druckansicht

Neue algorithmische Schranken für Isomorphieprobleme über Graphen und andere algebraische Strukturen

Fachliche Zuordnung Theoretische Informatik
Förderung Förderung von 2013 bis 2018
Projektkennung Deutsche Forschungsgemeinschaft (DFG) - Projektnummer 225911997
 
Erstellungsjahr 2018

Zusammenfassung der Projektergebnisse

The complexity of the graph isomorphism problem has been studied for a long time. This problem is very special since it is one of the few problems in NP that is not known to be NP-hard nor solvable in polynomial time. Recently several important results, most remarkably a quasi-polynomial time algorithm for the problem from Babai, have renewed the attention towards this problem and motivated new research. In this project we have approached the graph isomorphism problem from different perspectives, improving several aspects connected with its complexity. The most important results achieved are: Connections between the graph isomorphism problem and the area of proof complexity, proving exponential lower bounds for the problem in resolution, which give a theoretical explanation on why the the use of modern SAT-solvers do not help in dealing with the isomorphism problem. Fixed parameterized algorithms for several problems related to the isomorphism of hyper-graphs, exploring the tight border in which these problems become hard. A classification of the complexity of the graph isomorphism problem on several models of succinct input graphs. A first approach to the classification of isomorphism problems on solution graphs of Boolean formulas, where we show that the complexity of the problem is related to the formula structure. These results show once more the richness of the graph isomorphism problem and its connections to many areas of complexity.

Projektbezogene Publikationen (Auswahl)

 
 

Zusatzinformationen

Textvergrößerung und Kontrastanpassung