Computer Science – Computational Complexity
Scientific paper
2009-12-24
Computer Science
Computational Complexity
Scientific paper
Let X be randomly chosen from {-1,1}^n, and let Y be randomly chosen from the standard spherical Gaussian on R^n. For any (possibly unbounded) polytope P formed by the intersection of k halfspaces, we prove that |Pr [X belongs to P] - Pr [Y belongs to P]| < log^{8/5}k * Delta, where Delta is a parameter that is small for polytopes formed by the intersection of "regular" halfspaces (i.e., halfspaces with low influence). The novelty of our invariance principle is the polylogarithmic dependence on k. Previously, only bounds that were at least linear in k were known. We give two important applications of our main result: (1) A polylogarithmic in k bound on the Boolean noise sensitivity of intersections of k "regular" halfspaces (previous work gave bounds linear in k). (2) A pseudorandom generator (PRG) with seed length O((log n)*poly(log k,1/delta)) that delta-fools all polytopes with k faces with respect to the Gaussian distribution. We also obtain PRGs with similar parameters that fool polytopes formed by intersection of regular halfspaces over the hypercube. Using our PRG constructions, we obtain the first deterministic quasi-polynomial time algorithms for approximately counting the number of solutions to a broad class of integer programs, including dense covering problems and contingency tables.
Harsha Prahladh
Klivans Adam
Meka Raghu
No associations
LandOfFree
An Invariance Principle for Polytopes does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.
If you have personal experience with An Invariance Principle for Polytopes, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and An Invariance Principle for Polytopes will most certainly appreciate the feedback.
Profile ID: LFWR-SCP-O-555950