The article discusses the use of Physics-Informed Neural Networks (PINNs) for solving partial differential equations (PDEs). However, PINNs can face difficulties when dealing with equations with rapidly changing solutions. To address these issues, the authors propose a new framework called Binary Structured Physics-Informed Neural Network (BsPINN) that uses a binary structured neural network (BsNN). BsPINNs are effective and efficient in capturing local features of solutions, essential for learning rapidly changing solutions. They show superior convergence speed and heightened accuracy compared to PINNs in numerical experiments solving various equations.
Publication date: 24 Jan 2024
Project Page: Not provided
Paper: https://arxiv.org/pdf/2401.12806
