Optimally weighted loss functions for solving PDEs with Neural Networks
Autor: | Anastasia Borovykh, Cornelis W. Oosterlee, Remco van der Meer |
---|---|
Přispěvatelé: | Centrum Wiskunde & Informatica, Amsterdam (CWI), The Netherlands |
Rok vydání: | 2020 |
Předmět: |
Partial differential equation
Artificial neural network Generalization Heuristic (computer science) Applied Mathematics Loss functional Numerical Analysis (math.NA) Convection–diffusion equation Poisson equation Measure (mathematics) Neural network Computational Mathematics Data point Approximation error FOS: Mathematics Applied mathematics Mathematics - Numerical Analysis QA Scaling High-dimensional problems Mathematics |
Zdroj: | Journal of Computational and Applied Mathematics, 405 |
ISSN: | 0377-0427 |
DOI: | 10.48550/arxiv.2002.06269 |
Popis: | Recent works have shown that deep neural networks can be employed to solve partial differential equations, giving rise to the framework of physics informed neural networks (Raissi et al., 2007). We introduce a generalization for these methods that manifests as a scaling parameter which balances the relative importance of the different constraints imposed by partial differential equations. A mathematical motivation of these generalized methods is provided, which shows that for linear and well-posed partial differential equations, the functional form is convex. We then derive a choice for the scaling parameter that is optimal with respect to a measure of relative error. Because this optimal choice relies on having full knowledge of analytical solutions, we also propose a heuristic method to approximate this optimal choice. The proposed methods are compared numerically to the original methods on a variety of model partial differential equations, with the number of data points being updated adaptively. For several problems, including high-dimensional PDEs the proposed methods are shown to significantly enhance accuracy. |
Databáze: | OpenAIRE |
Externí odkaz: |