Evolutionary PINN Learning Algorithms Inspired by Approximation to Pareto Front for Solving Ill-Posed Problems

The article presents the development of new physics-informed evolutionary neural network learning algorithms. These algorithms aim to address the challenges of ill-posed problems by constructing a population close to the Pareto front. The study focuses on comparing the algorithm’s capabilities based...

Full description

Saved in:
Bibliographic Details
Published inComputation Vol. 11; no. 8; p. 166
Main Authors Lazovskaya, Tatiana, Tarkhov, Dmitriy, Chistyakova, Maria, Razumov, Egor, Sergeeva, Anna, Shemyakina, Tatiana
Format Journal Article
LanguageEnglish
Published Basel MDPI AG 01.08.2023
Subjects
Algorithms
Approximation
Boundary conditions
Boundary value problems
Data mining
discontinuous boundary conditions
Evolutionary algorithms
Genetic algorithms
Ill posed problems
Laplace equation
Machine learning
Mathematical models
Mathematical optimization
multi-criteria
multi-objective optimisation
Neural networks
pareto front
Partial differential equations
Performance evaluation
Physics
physics-informed neural networks
Smoothness
Russia
Online AccessGet full text

Cover

More Information
Summary:The article presents the development of new physics-informed evolutionary neural network learning algorithms. These algorithms aim to address the challenges of ill-posed problems by constructing a population close to the Pareto front. The study focuses on comparing the algorithm’s capabilities based on three quality criteria of solutions. To evaluate the algorithms’ performance, two benchmark problems have been used. The first involved solving the Laplace equation in square regions with discontinuous boundary conditions. The second problem considered the absence of boundary conditions but with the presence of measurements. Additionally, the study investigates the influence of hyperparameters on the final results. Comparisons have been made between the proposed algorithms and standard algorithms for constructing neural networks based on physics (commonly referred to as vanilla’s algorithms). The results demonstrate the advantage of the proposed algorithms in achieving better performance when solving incorrectly posed problems. Furthermore, the proposed algorithms have the ability to identify specific solutions with the desired smoothness.
ISSN:2079-3197
2079-3197
DOI:10.3390/computation11080166