Title
Complexity And Polymorphisms For Digraph Constraint Problems Under Some Basic Constructions
Abstract
The role of polymorphisms in determining the complexity of constraint satisfaction problems is well established. In this context, we study the stability of CSP complexity and polymorphism properties under some basic graph theoretic constructions. As applications we observe a collapse in the applicability of algorithms for CSPs over directed graphs with both a total source and a total sink: the corresponding CSP is solvable by the "few subpowers algorithm" if and only if it is solvable by a local consistency check algorithm. Moreover, we find that the property of "strict width" and solvability by few subpowers are unstable under first-order reductions. The analysis also yields a complete characterization of the main polymorphism properties for digraphs whose symmetric closure is a complete graph.
Year
DOI
Venue
2016
10.1142/S0218196716500600
INTERNATIONAL JOURNAL OF ALGEBRA AND COMPUTATION
Keywords
Field
DocType
Polymorphism, directed graph, one point extension, constraint satisfaction problem, homomorphism problem
Complete graph,Discrete mathematics,Local consistency,Combinatorics,Constraint graph,Directed graph,Constraint satisfaction problem,Complexity of constraint satisfaction,Symmetric closure,Constraint satisfaction dual problem,Mathematics
Journal
Volume
Issue
ISSN
26
7
0218-1967
Citations 
PageRank 
References 
2
0.37
16
Authors
3
Name
Order
Citations
PageRank
Marcel Jackson16211.86
Tomasz Kowalski212424.06
Todd Niven31016.11