Hybrid models of opinion dynamics with opinion-dependent connectivity

Paolo Frasca 1 Sophie Tarbouriech 2 Luca Zaccarian 2, 3
1 NECS-POST - Systèmes Commandés en Réseau
GIPSA-DA - Département Automatique, Inria Grenoble - Rhône-Alpes
2 LAAS-MAC - Équipe Méthodes et Algorithmes en Commande
LAAS - Laboratoire d'analyse et d'architecture des systèmes
Abstract : This paper is a first attempt at using tools from the theory of hybrid systems to study opinion dynamics on networks with opinion-dependent connectivity. According to the hybrid framework, our dynamics are represented by the combination of continuous flow dynamics and discrete jump dynamics. The flow embodies the attractive forces between the agents and is defined by an ordinary differential equation whose right-hand side is a Laplacian, whereas the jumps describe the activation or deactivation of the pairwise interactions between agents. We first reformulate the classical Hegselmann–Krause model in this framework and then define a novel interaction model, which has the property of being scale-invariant. We study the stability and convergence properties of both models by a Lyapunov analysis, showing convergence and clusterization of opinions.
Complete list of metadatas

Contributor : Paolo Frasca <>
Submitted on : Friday, November 30, 2018 - 9:48:19 AM
Last modification on : Saturday, October 26, 2019 - 1:34:03 AM
Long-term archiving on : Friday, March 1, 2019 - 1:18:47 PM


Files produced by the author(s)



Paolo Frasca, Sophie Tarbouriech, Luca Zaccarian. Hybrid models of opinion dynamics with opinion-dependent connectivity. Automatica, Elsevier, 2019, 100, pp.153-161. ⟨10.1016/j.automatica.2018.11.017⟩. ⟨hal-01940187⟩



Record views


Files downloads