Persistence in a large network of sparsely interacting neurons (Q2110336)

The main structure of the paper is as follows: first, the authors study the large time behavior of the finite network, then characterize the mean field limit process which is nonlinear process and showed its existence and uniqueness. They proved that there is a simple phase transition in terms of the parameters of the model: if the transmission of potential is strong enough (in terms of strength and frequency), then the limiting process is active forever; otherwise, it decays to zero. Then they studied propagation of chaos. They showed that the convergence, for fixed time, of the empirical measure of the finite network towards its mean-field limit, when the number of neurons grows to infinity. The article presents a biological neural network model driven by inhomogeneous Poisson processes accounting for the intrinsic randomness of synapses. The main novelty is the introduction of sparse interactions: each firing neuron triggers an instantaneous increase in electric potential to a fixed number of randomly chosen neurons. It is shwon that, as the number of neurons approaches infinity, the finite network converges to a nonlinear mean-field process characterized by a jump-type stochastic differential equation. This process displays a phase transition: the activity of a typical neuron in the infinite network either rapidly dies out, or persists forever, depending on the global parameters describing the intensity of interconnection. This provides a way to understand the emergence of persistent activity triggered by weak input signals in large neural networks.