|
Home
News
Citegeist
|
Browse Groups
Search Groups
Journals
|
FAQs
Howto
Discussion
|
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
A sharp threshold for a random constraint satisfaction problem |
Reviews
[Write a review of this article]
Find related articles from these CiteULike users
Find related articles with these CiteULike tags
Posting HistoryNEW
AbstractWe consider random instances I of a constraint satisfaction problem generalizing k-SAT: given a set of ordered k-tuples over n literals, and a set of q "bad" clause assignments, find an assignment which does not set any of the k-tuples to a bad clause assignment. We consider the case where k=[Omega](log n), and study the probability of satisfiability for a random instance I formed by including every k-tuple of literals independently with probability p. Appropriate choice of the bad clause assignments results in random instances of k-SAT and not-all-equal k-SAT. A second moment method calculation yields the sharp threshold
BibTeX record
RIS record