About optimal loss function for training physics-informed neural networks under respecting causality
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2023
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917973045805056 |
|---|---|
| author | Es'kin, Vasiliy A. Davydov, Danil V. Egorova, Ekaterina D. Malkhanov, Alexey O. Akhukov, Mikhail A. Smorkalov, Mikhail E. |
| author_facet | Es'kin, Vasiliy A. Davydov, Danil V. Egorova, Ekaterina D. Malkhanov, Alexey O. Akhukov, Mikhail A. Smorkalov, Mikhail E. |
| contents | A method is presented that allows to reduce a problem described by differential equations with initial and boundary conditions to the problem described only by differential equations. The advantage of using the modified problem for physics-informed neural networks (PINNs) methodology is that it becomes possible to represent the loss function in the form of a single term associated with differential equations, thus eliminating the need to tune the scaling coefficients for the terms related to boundary and initial conditions. The weighted loss functions respecting causality were modified and new weighted loss functions based on generalized functions are derived. Numerical experiments have been carried out for a number of problems, demonstrating the accuracy of the proposed methods. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2304_02282 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | About optimal loss function for training physics-informed neural networks under respecting causality Es'kin, Vasiliy A. Davydov, Danil V. Egorova, Ekaterina D. Malkhanov, Alexey O. Akhukov, Mikhail A. Smorkalov, Mikhail E. Numerical Analysis Artificial Intelligence Computational Physics 68T07 (Primary) 65Z05, 65M99 (Secondary) I.2.1; I.2.7; J.2 A method is presented that allows to reduce a problem described by differential equations with initial and boundary conditions to the problem described only by differential equations. The advantage of using the modified problem for physics-informed neural networks (PINNs) methodology is that it becomes possible to represent the loss function in the form of a single term associated with differential equations, thus eliminating the need to tune the scaling coefficients for the terms related to boundary and initial conditions. The weighted loss functions respecting causality were modified and new weighted loss functions based on generalized functions are derived. Numerical experiments have been carried out for a number of problems, demonstrating the accuracy of the proposed methods. |
| title | About optimal loss function for training physics-informed neural networks under respecting causality |
| topic | Numerical Analysis Artificial Intelligence Computational Physics 68T07 (Primary) 65Z05, 65M99 (Secondary) I.2.1; I.2.7; J.2 |
| url | https://arxiv.org/abs/2304.02282 |