Advanced Physics-Informed Neural Network with Residuals for Solving Complex Integral Equations
Mahdi Movahedian Moghaddam, Kourosh Parand, Saeed Reza Kheradpisheh · 2025
Read Advanced Physics-Informed Neural Network with Residuals for Solving Complex Integral Equations by Mahdi Movahedian Moghaddam, Kourosh Parand, Saeed Reza Kheradpisheh (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2501-16370.txt.
Description
Mahdi Movahedian Moghaddam, Kourosh Parand, and Saeed Reza Kheradpisheh published a 2025 paper on a neural network that solves integral equations more accurately. Such equations, where the unknown appears inside an integral, show up across physics and engineering. A popular approach trains a network by penalizing how badly it misses the equation. Those networks, though, can struggle with accuracy and with deep designs.
Their network, RISN, adds residual connections, the shortcuts that let very deep networks train well. It also builds in precise numerical tools for the integrals and for fractional derivatives. The design covers a wide range: single and multi-dimensional problems, systems of equations, and equations with wavy kernels. Across many tests, RISN beat both classic networks of this kind and two improved variants. Its errors were markedly lower. The authors offer it as a sturdier solver for problems where classic methods strain.
Catalog record
| CREATOR | Mahdi Movahedian Moghaddam, Kourosh Parand, Saeed Reza Kheradpisheh |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 67 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2501-16370