A hybrid semi-analytical and neural network approach for solving nonlinear Kolmogorov and Rosenau-Hyman equations
This paper presents a hybrid semi-analytical and physics-informed neural network framework for solving linear and nonlinear partial differential equations. The proposed approach combines the New Iterative Method (NIM) with Physics-Informed Neural Networks (PINNs): a low-order analytical approximation is generated by NI...