Table of Contents
Fetching ...

Hardness of the Binary Covering Radius Problem in Large $\ell_p$ Norms

Huck Bennett, Peter Ly

Abstract

We study the hardness of the $γ$-approximate decisional Covering Radius Problem on lattices in the $\ell_p$ norm ($γ$-$\text{GapCRP}_p$). Specifically, we prove that there is an explicit function $γ(p)$, with $γ(p) > 1$ for $p > p_0 \approx 35.31$ and $\lim_{p \to \infty} γ(p) = 9/8$, such that for any constant $\varepsilon > 0$, $(γ(p) - \varepsilon)$-$\text{GapCRP}_p$ is $\mathsf{NP}$-hard. This shows the first hardness of $\text{GapCRP}_p$ for explicit $p < \infty$. Work of Haviv and Regev (CCC, 2006 and CJTCS, 2012) previously showed $Π_2$-hardness of approximation for $\text{GapCRP}_p$ for all sufficiently large (but non-explicit) finite $p$ and for $p = \infty$. In fact, our hardness results hold for a variant of $\text{GapCRP}$ called the Binary Covering Radius Problem ($\text{BinGapCRP}$), which trivially reduces to both $\text{GapCRP}$ and the decisional Linear Discrepancy Problem ($\text{LinDisc}$) in any norm in an approximation-preserving way. We also show $Π_2$-hardness of $(9/8 - \varepsilon)$-$\text{BinGapCRP}$ in the $\ell_{\infty}$ norm for any constant $\varepsilon > 0$. Our work extends and heavily uses the work of Manurangsi (IPL, 2021), which showed $Π_2$-hardness of $(9/8 - \varepsilon)$-$\text{LinDisc}$ in the $\ell_{\infty}$ norm.

Hardness of the Binary Covering Radius Problem in Large $\ell_p$ Norms

Abstract

We study the hardness of the -approximate decisional Covering Radius Problem on lattices in the norm (-). Specifically, we prove that there is an explicit function , with for and , such that for any constant , - is -hard. This shows the first hardness of for explicit . Work of Haviv and Regev (CCC, 2006 and CJTCS, 2012) previously showed -hardness of approximation for for all sufficiently large (but non-explicit) finite and for . In fact, our hardness results hold for a variant of called the Binary Covering Radius Problem (), which trivially reduces to both and the decisional Linear Discrepancy Problem () in any norm in an approximation-preserving way. We also show -hardness of - in the norm for any constant . Our work extends and heavily uses the work of Manurangsi (IPL, 2021), which showed -hardness of - in the norm.
Paper Structure (21 sections, 18 theorems, 52 equations, 1 figure)

This paper contains 21 sections, 18 theorems, 52 equations, 1 figure.

Key Result

Theorem 1.1

For all $p > p_0 := 35.310188\ldots$ there exists an explicit value $\gamma(p) > 1$, specified in eq:np-hardness-factor, such that for any constant $\eps > 0$, $(\gamma(p) - \eps)$-$\problem{BinGapCRP}_p$ is $\NP$-hard. Furthermore, the values $\gamma(p)$ satisfy $\lim_{p \to \infty} \gamma(p) = 9/8

Figures (1)

  • Figure 1: A linear-log plot of the approximation factor $\gamma = \gamma(p)$ for which we show $\NP$-hardness of $(\gamma - \eps)$-$\problem{BinGapCRP}$ in the $\ell_p$ norm (in blue; see \ref{['thm:intro-np-hardness']}). The function $\gamma(p)$, specified in \ref{['eq:np-hardness-factor']}, is greater than $1$ for all $p > p_0 = 35.310188...$ and satisfies $\lim_{p \to \infty} \gamma(p) = 9/8$. (The blue dot is at the point $(p, \gamma) = (p_0, 1)$, where $p_0$ is the unique value of $p$ such that $\gamma(p) = 1$. The horizontal asymptote of $9/8$ is shown in red.) This factor of $9/8$ also appears in our $\Pi_2$-hardness result for $\problem{BinGapCRP}_{\infty}$ (\ref{['thm:intro-pi2-hardness']}) and in the work of Manurangsi manurangsi21 on $\problem{LinDisc}$, which is the foundation of our work.

Theorems & Definitions (42)

  • Theorem 1.1: $\NP$-hardness of approximation for $\problem{BinGapCRP}_p$
  • Theorem 1.2: $\Pi_2$-hardness of approximation for $\problem{BinGapCRP}_{\infty}$
  • Example 2.1
  • Definition 2.2
  • Definition 2.3
  • Definition 2.4
  • Definition 2.6: Constraint Satisfaction Problems (CSP)
  • Definition 2.7: $(\delta, \eps)$-$\text{NAE-E$k$-SAT}$ problem
  • Theorem 2.8
  • Lemma 3.1
  • ...and 32 more