Zanin M, Papo D, Sendiña-Nadal I, Boccaletti S
Center for Biomedical Technology, Technical University of Madrid, Campus Montegancedo, 28223 Pozuelo de Alarcón, Madrid, Spain.
Phys Rev E Stat Nonlin Soft Matter Phys. 2011 Dec;84(6 Pt 1):060102. doi: 10.1103/PhysRevE.84.060102. Epub 2011 Dec 20.
We report on the spontaneous emergence of computation from adaptive synchronization of networked dynamical systems. The fundamentals are nonlinear elements, interacting in a directed graph via a coupling that adapts itself to the synchronization level between two input signals. These units can emulate different Boolean logics, and perform any computational task in a Turing sense, each specific operation being associated with a given network's motif. The resilience of the computation against noise is proven, and the general applicability is demonstrated with regard to periodic and chaotic oscillators, and excitable systems mimicking neural dynamics.
我们报告了网络动力系统自适应同步中计算的自发出现。其基本要素是非线性元件，它们通过一种能根据两个输入信号之间的同步水平进行自我调整的耦合，在有向图中相互作用。这些单元可以模拟不同的布尔逻辑，并在图灵意义上执行任何计算任务，每个特定操作都与给定网络的基序相关联。证明了计算对噪声的鲁棒性，并针对周期和混沌振荡器以及模仿神经动力学的可激发系统展示了其普遍适用性。
