# Tribes

## Contents

## Definition

Let [math]n = ws[/math]. The tribes function with [math]s[/math] tribes of width [math]w[/math] is the boolean function [math]\mathrm{Tribes}_{w,s} : \{-1 ,+1\}^{ws} \rightarrow \{-1,+1\}[/math] defined by the DNF formula

[math]\mathrm{Tribes}_{w,s}(x) = \bigvee_{i=1}^{s} \bigwedge_{j=1}^{w} x_{i,j}[/math],

where we identify [math]-1[/math] with logical True and [math]+1[/math] with logical False.

The Tribes function corresponds to the follows voting scheme. There are [math]s[/math] tribes, each consists of [math]w[/math] members and the value of [math]x_{i,j}[/math] corresponds to the vote of the j-th member from the i-th tribe (with the above identification of logical values). First, each tribe takes a unanimous vote. Namely, if all tribe members voted True then the tribe vote is counted as True, otherwise the tribe vote is considered False. Then, if at least tribe voted True then the entire vote is considered True and otherwise it is considered False. In other words, the value of [math]\mathrm{Tribes}_{w,s}(x)[/math] is True if and only if at all the members of at least one tribe voted True.

The Tribes function is also known by the name of "read-once CNF".

## Properties

- Bias: When choosing an input [math]x[/math] uniformly at random, [math]\mathrm{Pr}_{x}[\mathrm{Tribes}_{w,s}(x) = -1] = 1 - (1-2^{-w})^{s}[/math].
- Fourier Expansion: Index the Fourier coefficients of [math]\mathrm{Tribes}_{w,s}(x)[/math] as follows. Given [math]T \subseteq [n][/math] write [math]T = (T_1,\ldots,T_s)[/math] where [math]T_i \subseteq \{(i-1)\cdot w+1,.\ldots , i\cdot w + w\}[/math] for [math]i = 1,2,\ldots,s[/math]. Then,

[math]\widehat{\mathrm{Tribes}_{w,s}}(T) = \begin{cases} 2(1-2^{-w})^{s}-1 & T = \emptyset\\ 2(-1)^{|\{i : T_i \neq \emptyset\}|+|T|}2^{-|\{i : T_i \neq \emptyset\}| \cdot w}(1-2^{-w})^{s-|\{i : T_i \neq \emptyset\}|} & T \neq \emptyset\\ \end{cases} [/math]

- Among almost asymptotically unbiased functions, Tribes asymptotically has smallest influence variables: One can choose the number of tribes and the tribe size so that the tribes function is almost an unbiased function, and with each variable having very little influence. Specifically, when [math]w = \log n - \log\log n + O(1)[/math] and [math]s = n / w[/math], we get that [math]\mathrm{Pr}_{x}[\mathrm{Tribes}_{w,s}(x) = -1] = 1/2 - O(\frac{\log n}{n})[/math] and the influence of every variable equals [math]\frac{\ln n}{n}(1+o(1))[/math]. This is the smallest maximum influence possible according to the KKL theorem
^{[1]}. - With the above choice of [math]w[/math] and [math]s[/math], the Tribes function is saturates (up to a constant factor) both the KKL and Talagrand's influence inequalities.