The Player's Effect

TL;DR

This paper investigates the influence of individual players on functions with inputs from multiple players, establishing bounds on the number of influential players and conditions under which players have positive effect.

Contribution

It provides a tight upper bound on the number of players with large effect under pairwise independence and explores conditions for the existence of players with positive effect.

Findings

Bound on the number of players with large effect under pairwise independence

Existence of small sets of players whose removal reduces effect

Conditions under which players have positive effect

Abstract

In a function that takes its inputs from various players, the effect of a player measures the variation he can cause in the expectation of that function. In this paper we prove a tight upper bound on the number of players with large effect, a bound that holds even when the players' inputs are only known to be pairwise independent. We also study the effect of a set of players, and show that there always exists a "small" set that, when eliminated, leaves every set with little effect. Finally, we ask whether there always exists a player with positive effect. We answer this question differently in various scenarios, depending on the properties of the function and the distribution of players' inputs. More specifically, we show that if the function is non-monotone or the distribution is only known to be pairwise independent, then it is possible that all players have 0 effect. If the distribution is pairwise independent with minimal support, on the other hand, then there must exist a player with "large" effect.