Title
Log-concavity and lower bounds for arithmetic circuits.
Abstract
One question that we investigate in this paper is, how can we build log-concave polynomials using sparse polynomials as building blocks? More precisely, let f = Sigma(d)(i=0) a(i),X-i E R+[X] be a polynomial satisfying the log-concavity condition a(i)(2), > Ta-i-1a(i),,+1 for every i is an element of {1,,d 1}, where r > 0. Whenever f can be written under the form f = Sigma(k)(i) = 1 Pi(m)(j)=1 f(i,j) where the polynomials fi have at most t monomials, it is clear that d ktm. Assuming that the have only non-negative coefficients, we improve this degree bound to d = O(km(2/3)t(2m/3)log(2/3) (kt)) if T > 1, and to d kmt if T = d(2d) This investigation has a complexity-theoretic motivation: we show that a suitable strengthening of the above results would imply a separation of the algebraic complexity classes VP and VNP. As they currently stand, these results are strong enough to provide a new example of a family of polynomials in VN P which cannot be computed by monotone arithmetic circuits of polynomial size.
Year
DOI
Venue
2015
10.1007/978-3-662-48054-0_30
Lecture Notes in Computer Science
Field
DocType
Volume
Discrete mathematics,Arithmetic circuits,Combinatorics,Newton polygon,Polynomial,Monomial,Mathematics,Bivariate polynomials
Journal
9235
ISSN
Citations 
PageRank 
0302-9743
0
0.34
References 
Authors
6
3
Name
Order
Citations
PageRank
Ignacio García-Marco162.66
Pascal Koiran2919113.85
Sébastien Tavenas3145.19