Abstract:Recently developed kinetic theory for neuronal network dynamics has been demonstrated to be a powerful theoretical framework for investigating coarse-grained dynamical properties of neuronal networks. The kinetic equations are a system of (1+1)dimensional nonlinear partial differential equations (PDE) on a bounded domain with the following features: (i) the boundary conditions are nonlinear and they are themselves a functional of the present solution; (ii) the PDEs themselves are self-consistently specified by parameters which are functions of the boundary values of the solution; and (iii) the PDEs can be stiff in space and time. Numerical methods are presented here for efficiently and accurately solving these kinetic equations. The essential ingredients in our numerical methods include (i) the system is discretized in time with an implicit Euler method within a spectral deferred correction framework, therefore, the PDEs of the kinetic theory are reduced to a sequence, in time, of boundary value problems (BVPs) with nonlinear boundary conditions; (ii) a set of auxiliary parameters is introduced to recast the original BVP with nonlinear boundary conditions as BVPs with linear boundary conditions – with additional algebraic constraints on the auxiliary parameters; (iii) A careful combination of two Newton’s iterates for the nonlinear BVP with linear boundary condition, interlaced with a Newton’s iterate for solving the associated algebraic constraints is constructed to achieve quadratic convergence for obtaining the solutions with self-consistent parameters. It is shown that a simple fixedpoint iteration can only achieve a linear convergence for the self-consistent parameters. The practicability and efficiency of our numerical methods for solving the kinetic equations are illustrated with numerical examples. It is further demonstrated that the kinetic equations can very well capture the coarse-grained dynamical properties of integrate-and-fire neuronal networks. Corresponding author: Aaditya V. Rangan, Courant Institute of Mathematical Sciences, New York University, New York, NY 10012. Phone: 212-998-3303, Fax: 212-995-4121, email: rangan@cims.nyu.edu.

Hyperbolic partial differential-difference equation in the mathematical modeling of neuronal firing and its numerical solution

Application of Efficient Numerical Methods in Solution of Ordinary Differential Equations for Modeling Electrical Activity in Cardiac Cells

Finite difference approximations for the first-order hyperbolic partial differential equation with point-wise delay

A model of neuronal bursting using three coupled first order differential equations

Approximate expressions of the bifurcating periodic solutions in a neuron model with delay-dependent parameters by perturbation approach

Simulating neuronal dynamics in fractional adaptive exponential integrate-and-fire models

Analytical solution for dynamic of neuronal populations

Numerical analysis of coupled dynamical biological networks: Modeling electrical information exchange among nerve cells using finite volume method

Towards the Modeling of Neuronal Firing by Gaussian Processes

A model for the periodic synaptic inhibition of a neuronal oscillator

A decision-making Fokker-Planck model in computational neuroscience

Numerical Methods for Solving Kinetic Equations of Neuronal Network Dynamics

Predicting the Firing Behavior of Neural Network Through the Bifurcation Analysis of Derivative Mean-Field Model

Qualitative Analysis of A Quadratic Integrate-And-Fire Neuron Model with State-Dependent Feedback Control

Mean-field method for generic conductance-based integrate-and-fire neurons with finite timescales

On a voltage-conductance kinetic system for integrate and fire neural networks

Mathematical Modeling and Analysis of Spatial Neuron Dynamics: Dendritic Integration and Beyond

Classical Solutions for a Nonlinear Fokker-Planck Equation Arising in Computational Neuroscience

Computational Fuzzy Modelling Approach to Analyze Neuronal Calcium Dynamics With Intracellular Fluxes

Well-Posedness and Regularity of Solutions to Neural Field Problems with Dendritic Processing

Qualitative properties of solutions for the noisy integrate and fire model in computational neuroscience