Identification of Biological Neural Network Using Jumping Gene Genetic Algorithm

In this paper, a jumping gene genetic algorithm is adopted to identify the topology of biological neural networks. The neural network is modeled with Hindmarsh-Rose neurons with synaptic coupling. Based on a single observable state of each neuron, it is possible to reveal the topology of the entire network under a framework of synchronization. The simulation results demonstrate that the topological structure of a neural network can be estimated accurately even if the exact models of the neurons are unknown.