Self-Organized Neural Network Structure Depending on the STDP Learning Rules

Abstract

Complex systems are widely studied in various fields of science. A neural network is one of the typical examples of the complex systems. Recent studies in neuroscience reported that the neural networks are dynamically self-organized by the spike-timing dependent synaptic plasticity (STDP). Although the neural networks change their structure using the STDP dynamically, neural networks are often analyzed in a static state. Thus, in this paper, we analyze neural network structure from a dynamical point of view. then, we show that the self-organized neural network to which the STDP learning rule is applied generates the small-world effect and randomness of the inter-spike intervals (ISIs) in the self-organized neural network increases as the small-world effect becomes higher.