### Abstract

In this paper we give evidence suggesting that MAX-CUT is NP-hard to approximate to within a factor of α _{GW}+ε, for all ε > 0, where α _{GW} denotes the approximation ratio achieved by the Goemans-Williamson algorithm [14], α _{GW} ≈ .878567. This result is conditional, relying on two conjectures: a) the Unique Games conjecture of Khot [24]; and, b) a very believable conjecture we call the Majority Is Stablest conjecture. These results indicate that the geometric nature of the Goemans-Williamson algorithm might be intrinsic to the MAX-CUT problem. The same two conjectures also imply that it is NP-hard to (β + ε)-approximate MAX-2SAT, where β ≈ .943943 is the minimum of (2 + 2/πθ)/(3 - cos(θ)) on (π/2, π). Motivated by our proof techniques, we show that if the MAX2CSP and MAX-2SAT problems are slightly restricted - in a way that seems to retain all their hardness - then they have (α _{GW} - ε )- and (β - ε)-approximation algorithms, respectively. Though we are unable to prove the Majority Is Stablest conjecture, we give some partial results and indicate possible directions of attack. Our partial results are enough to imply that MAX-CUT is hard to (3/4 + 1/2π + ε)-approximate (≈ .909155), assuming only the Unique Games conjecture. We also discuss MAX-2CSP problems over non-boolean domains and state some related results and conjectures. We show, for example, that the Unique Games conjecture implies that it is hard to approximate MAX-2LIN(q) to within any constant factor.

Original language | English (US) |
---|---|

Pages (from-to) | 146-154 |

Number of pages | 9 |

Journal | Proceedings - Annual IEEE Symposium on Foundations of Computer Science, FOCS |

State | Published - 2004 |

Event | Proceedings - 45th Annual IEEE Symposium on Foundations of Computer Science, FOCS 2004 - Rome, Italy Duration: Oct 17 2004 → Oct 19 2004 |

### ASJC Scopus subject areas

- Engineering(all)

## Fingerprint Dive into the research topics of 'Optimal inapproximability results for MAX-CUT and other 2-variable CSPs?'. Together they form a unique fingerprint.

## Cite this

*Proceedings - Annual IEEE Symposium on Foundations of Computer Science, FOCS*, 146-154.