Efficient system-wide coordination in noisy environments

Abstract

Many natural and social systems display global organization and coordination without centralized control. The origin of this global coordination is a topic of great current interest. Here we investigate a density-classification task as a model system for coordination and information processing in decentralized systems. We show that sophisticated strategies, selected under idealized conditions, are not robust to environmental changes. We also demonstrate that a simple heuristic is able to successfully complete the classification task under a broad range of environmental conditions. Our findings hint at the possibility that complex networks and ecologically efficient rules coevolve over time.

Similar impact of topological and dynamic noise on complex patterns

“…There, a self-organized critical state of the system is achieved without the introduction of dynamic noise (namely the occurrence of lightning strikes) by rewiring a small fraction of links. On the other hand, topological and dynamic noise can also play different functional roles in network dynamics, as reported in [8] for the case of a simple density classi- In the upper picture, a simple directed link has been rewired and accordingly, the states of a distant node j affect the update of node i through this shortcut in every time step t, t + 1, t + 2, . .…”

“…Pattern complexity is usually discussed in terms of the four Wolfram classes [6] distinguishing fixed points, periodic, chaotic and long-range complex behavior, respectively. Recently this approach to pattern formation has been linked with concepts from graph theory by studying the behavior of CA under variation of the underlying topology [7,8]. Investigations elucidating the link between dynamics and topology in complex networks are needed to ultimately understand the evolutionary building principles of biological networks.…”

“…Investigations elucidating the link between dynamics and topology in complex networks are needed to ultimately understand the evolutionary building principles of biological networks. The incorporation of noise or disorder in such systems is a natural step towards realistic models [8,9]. Qualitatively speaking, there are two possible ways of introducing disorder: On the level of signals and on the level of architecture.…”

“…In addition, however, topological perturbations can distribute global information among the local neighborhoods. Recently, simple binary dynamics in a noisy environment proved to be an excellent example for efficient information processing in decentralized systems [8].…”

Regularizing capacity of metabolic networks

“…Most of the rules lead to trivial spatio-temporal patterns, e.g. oscillations or a steady state like the frequently used majority rule [29,30]. Here, we choose one of those rules, which leads to non steady-state behavior on metabolic network topologies.…”