no code implementations • 13 Sep 2023 • Yiran Wang, Suchuan Dong
With the second method the high-dimensional PDE problem is reformulated through a constrained expression based on an Approximate variant of the Theory of Functional Connections (A-TFC), which avoids the exponential growth in the number of terms of TFC as the dimension increases.
no code implementations • 22 Mar 2023 • Yanxia Qian, Yongchao Zhang, Yunqing Huang, Suchuan Dong
Our analyses show that, with feed-forward neural networks having two hidden layers and the $\tanh$ activation function, the PINN approximation errors for the solution field, its time derivative and its gradient field can be effectively bounded by the training loss and the number of training data points (quadrature points).
no code implementations • 9 Oct 2022 • Suchuan Dong, Yiran Wang
The presented method has been compared with the physics-informed neural network method.
no code implementations • 24 Apr 2022 • Naxian Ni, Suchuan Dong
The HLConcELM method can produce highly accurate solutions to linear/nonlinear PDEs when the last hidden layer of the network is narrow and when it is wide.
no code implementations • 24 Jan 2022 • Suchuan Dong, Jielin Yang
For linear PDEs, enforcing the boundary/initial value problem on the collocation points leads to a separable nonlinear least squares problem about the network coefficients.
no code implementations • 27 Oct 2021 • Suchuan Dong, Jielin Yang
We consider the use of extreme learning machines (ELM) for computational partial differential equations (PDE).
no code implementations • 14 Mar 2021 • Suchuan Dong, Zongwei Li
In this paper we present a modified batch intrinsic plasticity (modBIP) method for pre-training the random coefficients in the ELM neural networks.
no code implementations • 4 Dec 2020 • Suchuan Dong, Zongwei Li
The computational performance of the current method is on par with, and oftentimes exceeds, the FEM performance.
no code implementations • 15 Jul 2020 • Suchuan Dong, Naxian Ni
We present a simple and effective method for representing periodic functions and enforcing exactly the periodic boundary conditions for solving differential equations with deep neural networks (DNN).