The Power of Social Information in Distributed Consensus in Ant-Colonies: Model and Analysis

Abstract

The decentralized cognition of animal groups is both a challenging biological problem and a potential basis for bio-inspired engineering design. The understanding of these systems and their application can benefit from modeling and analysis of the underlying algorithms. In Chapter 2, we define a modeling framework that can be used to formally represent all components of such algorithms. As an example application of the framework, we adapt to it the much-studied house-hunting algorithm used by emigrating colonies of Temnothorax ants to reach consensus on a new nest. We provide a Python simulator that encodes accurate individual behavior rules and produces simulated behaviors consistent with empirical observations, on both the individual and group levels. We use the simulator to make predictions about several aspects of collective emigration behavior, some with empirical support and some are new predictions. Critically, our results highlight the value of individual sensitivity to site population in ensuring consensus, and suggest its empirical measurement.

Though the above model captures a wide range of observed phenomenon and make new predictions, our work and previous work have mostly focused on experimental or modeling work, and lack rigorous mathematical justification. Building a theoretical understanding of the key mechanisms in the house-hunting process is necessary for the designs of novel distributed consensus algorithms. In order to do so, in this chapter we further simplified the model introduced in Chapter 2 and investigated the marginal benefits of the quorum sensing mechanism. We show theoretical confirmation of the hypothesized evolutionary advantage of quorum sensing in helping consensus. In addition, the desirable values of the quorum size from our theoretical results have been observed empirically.

It is our hope that the scientific insights and the modeling and mathematical tools can inspire further research from both the biology and computer science community.