SIAM Journal on Applied Dynamical Systems, Volume 19, Issue 1, Page 124-150, January 2020.
It has been recognized that many complex dynamical systems in the real world require a description in terms of multiplex networks, where a set of common, mutually connected nodes belong to distinct network layers and play a different role in each layer. In spite of recent progress toward data based inference of single-layer networks, to reconstruct complex systems with a multiplex structure remains largely open. In this paper, we articulate a mean-field based maximum likelihood estimation framework to address this problem. In a concrete manner, we reconstruct a class of prototypical duplex network systems hosting two categories of spreading dynamics, and we show that the structures of both layers can be simultaneously reconstructed from time series data. In addition to validating the framework using empirical and synthetic duplex networks, we carry out a detailed analysis to elucidate the impacts of network and dynamics parameters on the reconstruction accuracy and the robustness.

中文翻译：

---

基于数据的双工网络重构

SIAM应用动力系统杂志，第19卷，第1期，第124-150页，2020年1月。
已经认识到，现实世界中许多复杂的动力学系统都需要对多路复用网络进行描述，其中一组公共的，相互连接的节点属于不同的网络层，并且在每一层中都扮演着不同的角色。尽管最近在单层网络的基于数据的推理方面取得了进展，但是利用多路复用结构重建复杂的系统仍然很大。在本文中，我们阐述了一种基于均值场的最大似然估计框架来解决此问题。以一种具体的方式，我们重建了一类原型的双工网络系统，该系统承载了两类扩展的动力学，并且表明可以从时间序列数据中同时重建这两层的结构。除了使用经验和综合双工网络验证框架之外，