Heterogeneous pair approximation for voter models on networks

¹  Dipartimento di Fisica, “Sapienza” Università di Roma - P.le Aldo Moro 2, I-00185 Roma, Italy, EU
²  S. Anna School of Advanced Studies - P.za Martiri della Libertà 33, I-56100 Pisa, Italy, EU
³  SMC, INFM-CNR and Dipartimento di Fisica, “Sapienza” Università di Roma - P.le Aldo Moro 2, I-00185 Roma, Italy, EU

Corresponding author: claudio.castellano@roma1.infn.it

Received: 3 April 2009
Accepted: 16 November 2009

Abstract

For models whose evolution takes place on a network it is often necessary to augment the mean-field approach by considering explicitly the degree dependence of average quantities (heterogeneous mean field). Here we introduce the degree dependence in the pair approximation (heterogeneous pair approximation) for analyzing voter models on uncorrelated networks. This approach gives an essentially exact description of the dynamics, correcting some inaccurate results of previous approaches. The heterogeneous pair approximation introduced here can be applied in full generality to many other processes on complex networks.