About Speakers Schedule INS
偏微分方程算法和应用研讨会

Solving bivariate kinetic equations for polymer diffusion using deep learning

Speaker

邓伟华 , 兰州大学

Time

23 Mar, 08:45 - 09:30

Abstract

We derive a class of backward stochastic differential equations (BSDEs) for infinite dimensionally coupled nonlinear parabolic partial differential equations, thereby extending the deep BSDE method. In addition, we model a class of polymer dynamics accompanied by polymerization and depolymerization reactions, and derive the corresponding Fokker-Planck equations and Feynman-Kac equations. Due to chemical reactions, the system exhibits a Brownian yet non-Gaussian phenomenon, and the resulting equations are infinitely dimensionally coupled. We solve these equations numerically through our new deep BSDE method, and also solve a class of high-dimensional nonlinear equations, which verifies the effectiveness and shows approximation accuracy of the algorithm.