Efficient size estimation and impossibility of termination in uniform dense population protocols

TL;DR

This paper presents a uniform protocol for estimating the size of dense population protocols in polylogarithmic time, but proves that such protocols cannot be made terminating in the absence of an initial leader, highlighting fundamental limitations.

Contribution

It introduces a uniform size estimation protocol with high probability in polylogarithmic time and proves the impossibility of termination for such protocols without an initial leader in dense configurations.

Findings

Protocol estimates log(n) within constant factors in O(log^2 n) time.

Uniform protocols for tasks requiring more than constant time cannot be terminating in dense initial configurations.

With an initial leader, the size estimation protocol can be made terminating with high probability.

Abstract

We study uniform population protocols: networks of anonymous agents whose pairwise interactions are chosen at random, where each agent uses an identical transition algorithm that does not depend on the population size $n$. Many existing polylog$(n)$ time protocols for leader election and majority computation are nonuniform: to operate correctly, they require all agents to be initialized with an approximate estimate of $n$ (specifically, the exact value $\lfloor \log n \rfloor$). Our first main result is a uniform protocol for calculating $\log(n) \pm O(1)$ with high probability in $O(\log^2 n)$ time and $O(\log^4 n)$ states ($O(\log \log n)$ bits of memory). The protocol is converging but not terminating: it does not signal when the estimate is close to the true value of $\log n$. If it could be made terminating, this would allow composition with protocols, such as those for leader election or majority, that require a size estimate initially, to make them uniform (though with a small probability of failure). We do show how our main protocol can be indirectly composed with others in a simple and elegant way, based on the leaderless phase clock, demonstrating that those protocols can in fact be made uniform. However, our second main result implies that the protocol cannot be made terminating, a consequence of a much stronger result: a uniform protocol for any task requiring more than constant time cannot be terminating even with probability bounded above 0, if infinitely many initial configurations are dense: any state present initially occupies $\Omega(n)$ agents. (In particular, no leader is allowed.) Crucially, the result holds no matter the memory or time permitted. Finally, we show that with an initial leader, our size-estimation protocol can be made terminating with high probability, with the same asymptotic time and space bounds.