Loading [MathJax]/jax/output/HTML-CSS/config.js
arrow
Volume 15, Issue 3
Solving the Inverse Source Problem of the Fractional Poisson Equation by MC-fPINNs

Rui Sheng, Peiying Wu, Jerry Zhijian Yang & Cheng Yuan

East Asian J. Appl. Math., 15 (2025), pp. 565-590.

Published online: 2025-06

Export citation
  • Abstract

In this paper, we effectively solve the inverse source problem of the fractional Poisson equation using a Monte Carlo sampling-based PINN method (MC-fPINN). We construct two neural networks $u_{NN} (x;θ)$ and $f_{NN}(x;ψ)$ to approximate the solution $u^∗ (x)$ and the forcing term $f^∗(x)$ of the fractional Poisson equation. To optimize these networks, we use the Monte Carlo sampling method and define a new loss function combining the measurement data and underlying physical model. Meanwhile, we present a comprehensive error analysis for this method, along with a prior rule to select the appropriate parameters of neural networks. Numerical examples demonstrate the great accuracy and robustness of the method in solving high-dimensional problems up to 10D, with various fractional orders and noise levels of the measurement data ranging from 1% to 10%.

  • AMS Subject Headings

68T07, 65M12, 62G05

  • Copyright

COPYRIGHT: © Global Science Press

  • Email address
  • BibTex
  • RIS
  • TXT
@Article{EAJAM-15-565, author = {Sheng , RuiWu , PeiyingYang , Jerry Zhijian and Yuan , Cheng}, title = {Solving the Inverse Source Problem of the Fractional Poisson Equation by MC-fPINNs}, journal = {East Asian Journal on Applied Mathematics}, year = {2025}, volume = {15}, number = {3}, pages = {565--590}, abstract = {

In this paper, we effectively solve the inverse source problem of the fractional Poisson equation using a Monte Carlo sampling-based PINN method (MC-fPINN). We construct two neural networks $u_{NN} (x;θ)$ and $f_{NN}(x;ψ)$ to approximate the solution $u^∗ (x)$ and the forcing term $f^∗(x)$ of the fractional Poisson equation. To optimize these networks, we use the Monte Carlo sampling method and define a new loss function combining the measurement data and underlying physical model. Meanwhile, we present a comprehensive error analysis for this method, along with a prior rule to select the appropriate parameters of neural networks. Numerical examples demonstrate the great accuracy and robustness of the method in solving high-dimensional problems up to 10D, with various fractional orders and noise levels of the measurement data ranging from 1% to 10%.

}, issn = {2079-7370}, doi = {https://doi.org/10.4208/eajam.2024-072.150824}, url = {http://global-sci.org/intro/article_detail/eajam/24156.html} }
TY - JOUR T1 - Solving the Inverse Source Problem of the Fractional Poisson Equation by MC-fPINNs AU - Sheng , Rui AU - Wu , Peiying AU - Yang , Jerry Zhijian AU - Yuan , Cheng JO - East Asian Journal on Applied Mathematics VL - 3 SP - 565 EP - 590 PY - 2025 DA - 2025/06 SN - 15 DO - http://doi.org/10.4208/eajam.2024-072.150824 UR - https://global-sci.org/intro/article_detail/eajam/24156.html KW - Fractional Poisson equation, MC-fPINN, error analysis, inverse source problem. AB -

In this paper, we effectively solve the inverse source problem of the fractional Poisson equation using a Monte Carlo sampling-based PINN method (MC-fPINN). We construct two neural networks $u_{NN} (x;θ)$ and $f_{NN}(x;ψ)$ to approximate the solution $u^∗ (x)$ and the forcing term $f^∗(x)$ of the fractional Poisson equation. To optimize these networks, we use the Monte Carlo sampling method and define a new loss function combining the measurement data and underlying physical model. Meanwhile, we present a comprehensive error analysis for this method, along with a prior rule to select the appropriate parameters of neural networks. Numerical examples demonstrate the great accuracy and robustness of the method in solving high-dimensional problems up to 10D, with various fractional orders and noise levels of the measurement data ranging from 1% to 10%.

Sheng , RuiWu , PeiyingYang , Jerry Zhijian and Yuan , Cheng. (2025). Solving the Inverse Source Problem of the Fractional Poisson Equation by MC-fPINNs. East Asian Journal on Applied Mathematics. 15 (3). 565-590. doi:10.4208/eajam.2024-072.150824
Copy to clipboard
The citation has been copied to your clipboard