A dichotomy theorem for nonuniform CSPs simplified
Computational Complexity
2020-07-20 v1 Logic in Computer Science
Abstract
In a non-uniform Constraint Satisfaction problem CSP(G), where G is a set of relations on a finite set A, the goal is to find an assignment of values to variables subject to constraints imposed on specified sets of variables using the relations from G. The Dichotomy Conjecture for the non-uniform CSP states that for every constraint language G the problem CSP(G) is either solvable in polynomial time or is NP-complete. It was proposed by Feder and Vardi in their seminal 1993 paper. In this paper we confirm the Dichotomy Conjecture.
Cite
@article{arxiv.2007.09099,
title = {A dichotomy theorem for nonuniform CSPs simplified},
author = {Andrei A. Bulatov},
journal= {arXiv preprint arXiv:2007.09099},
year = {2020}
}
Comments
This is an updated and improved version of the proof of the CSP dichotomy from CoRR abs/1703.03021, 2017. arXiv admin note: text overlap with arXiv:1703.03021