Revision #1 Authors: Alexander Golovnev, Gleb Posobin, Oded Regev, Omri Weinstein

Accepted on: 8th December 2021 19:11

Downloads: 422

Keywords:

Proving super-logarithmic data structure lower bounds in the static group model has been a fundamental challenge in computational geometry since the early 80's. We prove a polynomial ($n^{\Omega(1)}$) lower bound for an explicit range counting problem of $n^3$ convex polygons in $\mathbb{R}^2$ (each with $n^{\tilde{O}(1)}$ facets/semialgebraic-complexity), against linear storage arithmetic data structures in the group model. Our construction and analysis are based on a combination of techniques in Diophantine approximation, pseudorandomness, and compressed sensingâ€”in particular, on the existence and partial derandomization of optimal binary compressed sensing matrices in the polynomial sparsity regime ($k = n^{1-\delta}$). As a byproduct, this establishes a (logarithmic) separation between compressed sensing matrices and the stronger RIP property.

Improved the presentation of the main results.

TR20-057 Authors: Alexander Golovnev, Gleb Posobin, Oded Regev, Omri Weinstein

Publication: 26th April 2020 14:44

Downloads: 1267

Keywords:

Proving super-logarithmic data structure lower bounds in the static \emph{group model} has been a fundamental challenge in computational geometry since the early 80's. We prove a polynomial ($n^{\Omega(1)}$) lower bound for an explicit range counting problem of $n^3$ convex polygons in $\R^2$ (each with $n^{\tilde{O}(1)}$ facets/semialgebraic-complexity), against linear storage arithmetic data structures in the group model. Our construction and analysis are based on a combination of techniques in Diophantine approximation, pseudorandomness, and compressed sensing---in particular, on the existence and partial derandomization of optimal \emph{binary} compressed sensing matrices in the polynomial sparsity regime ($k = n^{1-\delta}$). As a byproduct, this establishes a (logarithmic) separation between compressed sensing matrices and the stronger RIP property.