Narrow proofs may be maximally long

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

We prove that there are 3-CNF formulas over n variables that can be refuted in resolution in width w but require resolution proofs of size nω(w). This shows that the simple counting argument that any formula refutable in width w must have a proof in size nO(ω) is essentially tight. Moreover, our lower bounds can be generalized to polynomial calculus resolution (PCR) and Sherali-Adams, implying that the corresponding size upper bounds in terms of degree and rank are tight as well. Our results do not extend all the way to Lasserre, however-the formulas we study have Lasserre proofs of constant rank and size polynomial in both n and w.

Original languageEnglish
Title of host publicationProceedings - IEEE 29th Conference on Computational Complexity, CCC 2014
Number of pages12
PublisherIEEE Computer Society Press
Publication date2014
Pages286-297
Article number6875497
ISBN (Print)9781479936267
DOIs
Publication statusPublished - 2014
Externally publishedYes
Event29th Annual IEEE Conference on Computational Complexity, CCC 2014 - Vancouver, BC, Canada
Duration: 11 Jun 201413 Jun 2014

Conference

Conference29th Annual IEEE Conference on Computational Complexity, CCC 2014
LandCanada
ByVancouver, BC
Periode11/06/201413/06/2014
SponsorCommittee on Mathematical, Foundations of Computing, IEEE Computer Society Technical
SeriesProceedings of the Annual IEEE Conference on Computational Complexity
ISSN1093-0159

    Research areas

  • degree, Lasserre, length, PCR, polynomial calculus, proof complexity, rank, resolution, Sherali-Adams, size, width

ID: 251870042