Title :
LS+ Lower Bounds from Pairwise Independence
Author :
Tulsiani, Madhur ; Worah, Pratik
Author_Institution :
TTI Chicago, Chicago, IL, USA
Abstract :
We consider the complexity of LS+ refutations of unsatisfiable instances of Constraint Satisfaction Problems (k-CSPs) when the underlying predicate supports a pairwise independent distribution on its satisfying assignments. This is the most general condition on the predicates under which the corresponding MAX k-CSP problem is known to be approximation resistant. We show that for random instances of such k-CSPs on n variables, even after Ω(n) rounds of the LS+ hierarchy, the integrality gap remains equal to the approximation ratio achieved by a random assignment. In particular, this also shows that LS+ refutations for such instances require rank Ω(n). We also show the stronger result that refutations for such instances in the static LS+ proof system requires size exp(Ω(n)).
Keywords :
computational complexity; constraint satisfaction problems; theorem proving; LS+ lower bounds; LS+ refutation complexity; MAX k-CSP problem; approximation ratio; constraint satisfaction problems; integrality gap; pairwise independent distribution; random assignment; static LS+ proof system; Approximation methods; Complexity theory; Context; Matrix decomposition; Polynomials; Resistance;
Conference_Titel :
Computational Complexity (CCC), 2013 IEEE Conference on
Conference_Location :
Stanford, CA
DOI :
10.1109/CCC.2013.21