Table of Contents
Fetching ...

Geometric Variants of the Gale--Berlekamp Switching Game

Adrian Dumitrescu

TL;DR

This work extends the Gale–Berlekamp switching game to arbitrary boards and switch budgets, deriving tight asymptotics for maximum signed discrepancy under a variety of settings. Using probabilistic methods and projections $u,v$ of boards, it shows $F(H)=\Theta(A^{3/4})$ for dense boards and provides precise constants in several special configurations, including inscribed squares and disks. It also analyzes upgrades and downgrades of the switch set, proving $F(H)=\Theta(n^{3/2} t^{1/2})$ with $t n$ extra switches and characterizing discrepancy under budgeted removals. Finally, the paper extends the framework to a hyperbola-bound planar board and a cubic 3D board, obtaining $\Theta(n)$– and $\Theta(n^{5/2})$ growth rates respectively, thus generalizing Brown–Spencer’s original $\Theta(n^{3/2})$ result and illustrating a unified approach to lattice-board discrepancy problems.

Abstract

The Gale-Berlekamp switching game is played on the following device: $G_n=\{1,2,\ldots,n\} \times \{1,2,\ldots,n\}$ is an $n \times n$ array of lights is controlled by $2n$ switches, one for each row or column. Given an (arbitrary) initial configuration of the board, the objective is to have as many lights on as possible. Denoting the maximum difference (discrepancy) between the number of lights that are on minus the number of lights that are off by $F(n)$, it is known (Brown and Spencer, 1971) that $F(n)= Θ(n^{3/2})$, and more precisely, that $F(n) \geq \left( 1+ o(1) \right) \sqrt{\frac{2}π} n^{3/2} \approx 0.797 \ldots n^{3/2}$. Here we extend the game to other playing boards. For example: (i)~For any constant $c>1$, if $c n$ switches are conveniently chosen, then the maximum discrepancy for the square board is $Ω(n^{3/2})$. From the other direction, suppose we fix any set of $a$ column switches, $b$ row switches, where $a \geq b$ and $a+b=n$. Then the maximum discrepancy is at most $-b (n-b)$. (ii) A board $H \subset \{1,\ldots,n\}^2$, with area $A=|H|$, is \emph{dense} if $A \geq c (u+v)^2$, for some constant $c>0$, where $u= |\{x \colon (x,y) \in H\}|$ and $v=|\{y \colon (x,y) \in H\}|$. For a dense board of area $A$, we show that the maximum discrepancy is $Θ(A^{3/4})$. This result is a generalization of the Brown and Spencer result for the original game. (iii) If $H$ consists of the elements of $G_n$ below the hyperbola $xy=n$, then its maximum discrepancy is $Ω(n)$ and $O(n (\log n)^{1/2})$.

Geometric Variants of the Gale--Berlekamp Switching Game

TL;DR

This work extends the Gale–Berlekamp switching game to arbitrary boards and switch budgets, deriving tight asymptotics for maximum signed discrepancy under a variety of settings. Using probabilistic methods and projections of boards, it shows for dense boards and provides precise constants in several special configurations, including inscribed squares and disks. It also analyzes upgrades and downgrades of the switch set, proving with extra switches and characterizing discrepancy under budgeted removals. Finally, the paper extends the framework to a hyperbola-bound planar board and a cubic 3D board, obtaining – and growth rates respectively, thus generalizing Brown–Spencer’s original result and illustrating a unified approach to lattice-board discrepancy problems.

Abstract

The Gale-Berlekamp switching game is played on the following device: is an array of lights is controlled by switches, one for each row or column. Given an (arbitrary) initial configuration of the board, the objective is to have as many lights on as possible. Denoting the maximum difference (discrepancy) between the number of lights that are on minus the number of lights that are off by , it is known (Brown and Spencer, 1971) that , and more precisely, that . Here we extend the game to other playing boards. For example: (i)~For any constant , if switches are conveniently chosen, then the maximum discrepancy for the square board is . From the other direction, suppose we fix any set of column switches, row switches, where and . Then the maximum discrepancy is at most . (ii) A board , with area , is \emph{dense} if , for some constant , where and . For a dense board of area , we show that the maximum discrepancy is . This result is a generalization of the Brown and Spencer result for the original game. (iii) If consists of the elements of below the hyperbola , then its maximum discrepancy is and .

Paper Structure

This paper contains 16 sections, 10 theorems, 16 equations, 3 figures, 1 table.

Key Result

Proposition 1

Let $H$ be an $X$-shaped board inscribed in $G_n$ (with $u=v=n$). Then its maximum discrepancy is $F(H)=n$.

Figures (3)

  • Figure 1: Left: a square board whose signed discrepancy is $5$. Center: an $X$-shaped board whose signed discrepancy is $-1$; its maximum discrepancy is $5$. Right: a rotated square board.
  • Figure 2: Switches on slanted lines of slope $t=3$ and horizontal lines.
  • Figure 3: A square board with $n=8$. Left: $a=6, b=3$ ($9$ switches). Center: $a=b=6$ ($12$ switches). Right: $a=6, b=2$ ($8$ switches).

Theorems & Definitions (10)

  • Proposition 1
  • Theorem 2
  • Theorem 3
  • Theorem 4
  • Theorem 5
  • Theorem 6
  • Theorem 7
  • Theorem 8
  • Theorem 9
  • Theorem 10