Data Clustering as an Emergent Consensus of Autonomous Agents

TL;DR

This paper introduces a novel data segmentation method inspired by collective behavior, using a density-induced consensus protocol, with rigorous analysis and applications to shape datasets and images, augmenting classical clustering techniques.

Contribution

It presents a mathematically rigorous consensus-based clustering algorithm that extends classical methods like DBSCAN, connecting data clustering with emergent collective behavior.

Findings

Effective segmentation of shape datasets and images

Theoretical analysis of the consensus model

Connection between clustering and collective behavior

Abstract

We present a data segmentation method based on a first-order density-induced consensus protocol. We provide a mathematically rigorous analysis of the consensus model leading to the stopping criteria of the data segmentation algorithm. To illustrate our method, the algorithm is applied to two-dimensional shape datasets and selected images from Berkeley Segmentation Dataset. The method can be seen as an augmentation of classical clustering techniques for multimodal feature space, such as DBSCAN. It showcases a curious connection between data clustering and collective behavior.