Ruling out ptas for graph min-bisection, dense k-subgraph, and bipartite clique

Research output: Contribution to journalArticle

Abstract

Assuming that NP ⊈ ∩ε > 0 BPTIME(2 ), we show that graph min-bisection, dense fc-subgraph, and bipartite clique have no polynomial time approximation scheme (PTAS). We give a reduction from the minimum distance of code (MDC) problem. Starting with an instance of MDC, we build a quasi-random probabilistically checkable proof (PCP) that suffices to prove the desired inapproximability results. In a quasi-random PCP, the query pattern of the verifier looks random in a certain precise sense. Among the several new techniques we introduce, the most interesting one gives a way of certifying that a given polynomial belongs to a given linear subspace of polynomials. As is important for our purpose, the certificate itself happens to be another polynomial, and it can be checked probabilistically by reading a constant number of its values.

Original languageEnglish (US)
Pages (from-to)1025-1071
Number of pages47
JournalSIAM Journal on Computing
Volume36
Issue number4
DOIs
StatePublished - 2006

Fingerprint

Bisection
Probabilistically Checkable Proofs
Clique
Subgraph
Polynomials
Minimum Distance
Polynomial
Graph in graph theory
Inapproximability
Polynomial Time Approximation Scheme
Certificate
Subspace
Query

Keywords

  • Approximation algorithms
  • Hardness of approximation
  • Probabilistically checkable proofs (PCPs)

ASJC Scopus subject areas

  • Computational Theory and Mathematics
  • Applied Mathematics
  • Theoretical Computer Science

Cite this

Ruling out ptas for graph min-bisection, dense k-subgraph, and bipartite clique. / Khot, Subhash.

In: SIAM Journal on Computing, Vol. 36, No. 4, 2006, p. 1025-1071.

Research output: Contribution to journalArticle

@article{4d2ec7d056d44d12a544fe593eb2786a,
title = "Ruling out ptas for graph min-bisection, dense k-subgraph, and bipartite clique",
abstract = "Assuming that NP ⊈ ∩ε > 0 BPTIME(2 nε), we show that graph min-bisection, dense fc-subgraph, and bipartite clique have no polynomial time approximation scheme (PTAS). We give a reduction from the minimum distance of code (MDC) problem. Starting with an instance of MDC, we build a quasi-random probabilistically checkable proof (PCP) that suffices to prove the desired inapproximability results. In a quasi-random PCP, the query pattern of the verifier looks random in a certain precise sense. Among the several new techniques we introduce, the most interesting one gives a way of certifying that a given polynomial belongs to a given linear subspace of polynomials. As is important for our purpose, the certificate itself happens to be another polynomial, and it can be checked probabilistically by reading a constant number of its values.",
keywords = "Approximation algorithms, Hardness of approximation, Probabilistically checkable proofs (PCPs)",
author = "Subhash Khot",
year = "2006",
doi = "10.1137/S0097539705447037",
language = "English (US)",
volume = "36",
pages = "1025--1071",
journal = "SIAM Journal on Computing",
issn = "0097-5397",
publisher = "Society for Industrial and Applied Mathematics Publications",
number = "4",

}

TY - JOUR

T1 - Ruling out ptas for graph min-bisection, dense k-subgraph, and bipartite clique

AU - Khot, Subhash

PY - 2006

Y1 - 2006

N2 - Assuming that NP ⊈ ∩ε > 0 BPTIME(2 nε), we show that graph min-bisection, dense fc-subgraph, and bipartite clique have no polynomial time approximation scheme (PTAS). We give a reduction from the minimum distance of code (MDC) problem. Starting with an instance of MDC, we build a quasi-random probabilistically checkable proof (PCP) that suffices to prove the desired inapproximability results. In a quasi-random PCP, the query pattern of the verifier looks random in a certain precise sense. Among the several new techniques we introduce, the most interesting one gives a way of certifying that a given polynomial belongs to a given linear subspace of polynomials. As is important for our purpose, the certificate itself happens to be another polynomial, and it can be checked probabilistically by reading a constant number of its values.

AB - Assuming that NP ⊈ ∩ε > 0 BPTIME(2 nε), we show that graph min-bisection, dense fc-subgraph, and bipartite clique have no polynomial time approximation scheme (PTAS). We give a reduction from the minimum distance of code (MDC) problem. Starting with an instance of MDC, we build a quasi-random probabilistically checkable proof (PCP) that suffices to prove the desired inapproximability results. In a quasi-random PCP, the query pattern of the verifier looks random in a certain precise sense. Among the several new techniques we introduce, the most interesting one gives a way of certifying that a given polynomial belongs to a given linear subspace of polynomials. As is important for our purpose, the certificate itself happens to be another polynomial, and it can be checked probabilistically by reading a constant number of its values.

KW - Approximation algorithms

KW - Hardness of approximation

KW - Probabilistically checkable proofs (PCPs)

UR - http://www.scopus.com/inward/record.url?scp=34547839125&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=34547839125&partnerID=8YFLogxK

U2 - 10.1137/S0097539705447037

DO - 10.1137/S0097539705447037

M3 - Article

AN - SCOPUS:34547839125

VL - 36

SP - 1025

EP - 1071

JO - SIAM Journal on Computing

JF - SIAM Journal on Computing

SN - 0097-5397

IS - 4

ER -