1. An artificial neural network approach to identify the parameter in a nonlinear subdiffusion model.
- Author
-
Oulmelk, A., Srati, M., Afraites, L., and Hadri, A.
- Subjects
- *
MATHEMATICAL analysis , *NONLINEAR equations - Abstract
In this paper, we propose an artificial neural network approach to identify the parameter in a non-linear subdiffusion model from additional data. Instead of determining the parameter in the time fractional diffusion model by its form itself, we approximate it in the form of an artificial neural network. The key point of this approach relies on the approximation capability of neural networks. We formulate this inverse problem as an optimal control one, and we demonstrate the existence of the solution for the control problem and provide a mathematical analysis and the derivation of optimal conditions. Moreover, various numerical tests of the regular and singular examples have shown that the artificial neural network method (ANN) is effective. This is reinforced by its numerical comparison with the gradient descent, the alternating direction multiplier method (ADMM), physics-informed neural network (PINN) and DeepONet method. • A new formulation of the inverse problem for a nonlinear subdiffusion model based on an artificial neural network. • Theoretical analysis of the inverse problem by optimal control formulation. • Proposition of an approach for solving the inverse problem by the gradient descent algorithm. • A qualitative and quantitative comparative study with existing methods. [ABSTRACT FROM AUTHOR]
- Published
- 2023
- Full Text
- View/download PDF