Reconstruction of chaotic neural network from observed firing rates

Abstract

Randomly coupled neural fields demonstrate chaotic variation of firing rates, if the coupling is strong enough, as has been shown by Sompolinsky et. al [Phys. Rev. Lett., v. 61, 259 (1988)]. We present a method for reconstruction of the coupling matrix from the observations of the chaotic firing rates. The approach is based on the particular property of the nonlinearity in the coupling, as the latter is determined by a sigmoidal gain function. We demonstrate that for a large enough data set, the method gives an accurate estimation of the coupling matrix and of other parameters of the system, including the gain function.