This preview shows pages 1–2. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: 6.841 Advanced Complexity Theory March 19, 2007 Lecture 12 Lecturer: Madhu Sudan Scribe: Suho Oh, Jose Soto 1 Overview Randomized Reductions. ValiantVazirani: SAT RP Unique SAT . Todas Theorem: PH P # P . 2 The Theorem of ValiantVazirani. To state this theorem we will need some definitions first: Definition 1 ( Unique SAT promise problem) . Unique SAT = ( U Y ES ,U NO ) . U Y ES = {  has 1 satisfying assignment } . U NO = {  has 0 satisfying assignment } . Definition 2 (Randomized Reductions) Given two promise problems = ( Y ES , NO ) and = ( Y ES , NO ) . We say that reduces to under a BP randomized reduction BP if there exists a probabilistic polynomial time algorithm A , a polynomial p ( n ) and a polynomial time computable function s ( n ) such that: x Y ES = A ( x ) Y ES w.p. s ( n ) + 1 p ( n ) . x NO = A ( x ) negationslash NO w.p. s ( n ) . [ A ( x ) NO w.p. 1 s ( n )] . When s ( n ) = 0 we say that it is a RP randomized reduction and we denote it by RP . Using the previous definition we can state the theorem as follows: Theorem 1 (ValiantVazirani) SAT RP Unique SAT. To find an RP reduction a natural idea is to map an instance ( x ) of SAT into a new formula ( x ) = ( x ) f ( x ), where f ( x ) is a sufficiently nice formula. In that way if ( x ) SAT NO then we would know that ( x ) has no satisfying assignment, and so ( x ) U NO . The problem is to determine a nice f ( x ) such that if SAT Y ES , then ( x ) has exactly one satisfying assignment with enough probability....
View
Full
Document
This note was uploaded on 04/02/2010 for the course CS 6.841 taught by Professor Madhusudan during the Spring '09 term at MIT.
 Spring '09
 MadhuSudan

Click to edit the document details