English

An optimal lower bound for monotonicity testing over hypergrids

Data Structures and Algorithms 2013-04-22 v1 Discrete Mathematics

Abstract

For positive integers n,dn, d, consider the hypergrid [n]d[n]^d with the coordinate-wise product partial ordering denoted by \prec. A function f:[n]dNf: [n]^d \mapsto \mathbb{N} is monotone if xy\forall x \prec y, f(x)f(y)f(x) \leq f(y). A function ff is \eps\eps-far from monotone if at least an \eps\eps-fraction of values must be changed to make ff monotone. Given a parameter \eps\eps, a \emph{monotonicity tester} must distinguish with high probability a monotone function from one that is \eps\eps-far. We prove that any (adaptive, two-sided) monotonicity tester for functions f:[n]dNf:[n]^d \mapsto \mathbb{N} must make Ω(\eps1dlogn\eps1log\eps1)\Omega(\eps^{-1}d\log n - \eps^{-1}\log \eps^{-1}) queries. Recent upper bounds show the existence of O(\eps1dlogn)O(\eps^{-1}d \log n) query monotonicity testers for hypergrids. This closes the question of monotonicity testing for hypergrids over arbitrary ranges. The previous best lower bound for general hypergrids was a non-adaptive bound of Ω(dlogn)\Omega(d \log n).

Keywords

Cite

@article{arxiv.1304.5264,
  title  = {An optimal lower bound for monotonicity testing over hypergrids},
  author = {Deeparnab Chakrabarty and C. Seshadhri},
  journal= {arXiv preprint arXiv:1304.5264},
  year   = {2013}
}
R2 v1 2026-06-22T00:02:39.490Z