Nondeterministic Polynomial Time versus Nondeterministic Logarithmic Space (1996)
| Venue: | In Proceedings, Twelfth Annual IEEE Conference on Computational Complexity |
| Citations: | 20 - 1 self |
BibTeX
@INPROCEEDINGS{Fortnow96nondeterministicpolynomial,
author = {Lance Fortnow},
title = {Nondeterministic Polynomial Time versus Nondeterministic Logarithmic Space},
booktitle = {In Proceedings, Twelfth Annual IEEE Conference on Computational Complexity},
year = {1996},
pages = {52--60},
publisher = {IEEE Computer Society Press}
}
Years of Citing Articles
Bookmark
 |
 |
 |
 |
 |
 |
OpenURL
Abstract
We discuss the possibility of using the relatively old technique of diagonalization to separate complexity classes, in particular NL from NP. We show several results in this direction. ffl Any nonconstant level of the polynomial-time hierarchy strictly contains NL. ffl SAT is not simultaneously in NL and deterministic n log j n time for any j. ffl On the negative side, we present a relativized world where P = NP but any nonconstant level of the polynomial-time hierarchy differs from P. 1 Introduction Separating complexity classes remains the most important and difficult of problems in theoretical computer science. Circuit complexity and other techniques on finite functions have seen some exciting early successes (see the survey of Boppana and Sipser [BS90]) but have yet to achieve their promise of separating complexity classes above logarithmic space. Other techniques based on logic and geometry also have given us separations only on very restricted models. We should turn back to...
