Deducing Local Rules for Solving Global Tasks with Random Boolean Networks

It has been shown that uniform as well as non-uniform cellular automata (CA) can be evolved to perform certain computational tasks. Random Boolean networks are a generalization of two-state cellular automata, where the interconnection topology and the cell's rules are specified at random. Here we present a novel analytical approach to find the local rules of random Boolean networks (RBNs) to solve the global density classification and the synchronization task from any initial configuration. We quantitatively and qualitatively compare our results with previously published work on cellular automata and show that randomly interconnected automata are computationally more efficient in solving these two global tasks. Our approach also provides convergence and quality estimates and allows the networks to be randomly rewired during operation, without affecting the global performance. Finally, we show that RBNs outperform small-world topologies on the density classification task and that they perform equally well on the synchronization task. Our novel approach and the results may have applications in designing robust complex networks and locally interacting distributed computing systems for solving global tasks.

Abstract licence: CC0

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.