Decentralized Clustering and Linking by Networked Agents

TL;DR

This paper introduces a decentralized clustering algorithm for multi-task networks where agents identify similar models and collaborate to improve inference, even with unknown models and cluster memberships.

Contribution

The paper presents a novel integrated learning and clustering algorithm for multi-task networks, with proven exponential decay of error probabilities and strategies to leverage network links for better performance.

Findings

Error probabilities decay exponentially with step-size

Algorithm effectively identifies clusters of similar objectives

Links between different objectives can be exploited for improved inference

Abstract

We consider the problem of decentralized clustering and estimation over multi-task networks, where agents infer and track different models of interest. The agents do not know beforehand which model is generating their own data. They also do not know which agents in their neighborhood belong to the same cluster. We propose a decentralized clustering algorithm aimed at identifying and forming clusters of agents of similar objectives, and at guiding cooperation to enhance the inference performance. One key feature of the proposed technique is the integration of the learning and clustering tasks into a single strategy. We analyze the performance of the procedure and show that the error probabilities of types I and II decay exponentially to zero with the step-size parameter. While links between agents following different objectives are ignored in the clustering process, we nevertheless show how to exploit these links to relay critical information across the network for enhanced performance. Simulation results illustrate the performance of the proposed method in comparison to other useful techniques.