Physics-informed neural networks with non-differentiable loss
Physics-informed Neural Networks (PINN) are special neural networks that are designed for scientific computing tasks. Recent research has found its promising capability to integrate any given law of physics in different forms including general nonlinear partial differentiable equations. It has sh...
Main Author: | Yang, Junyan |
---|---|
Other Authors: | Mao Kezhi |
Format: | Final Year Project (FYP) |
Language: | English |
Published: |
Nanyang Technological University
2022
|
Subjects: | |
Online Access: | https://hdl.handle.net/10356/158021 |
Similar Items
-
CAN-PINN: a fast physics-informed neural network based on coupled-automatic-numerical differentiation method
by: Chiu, Pao-Hsiung, et al.
Published: (2022) -
Fourier warm start for physics-informed neural networks
by: Jin, Ge, et al.
Published: (2024) -
Information extraction with neural networks
by: Lee, Ji Young, Ph. D. Massachusetts Institute of Technology
Published: (2017) -
Neural network model for differential GPS
by: Low, Kwong Hwee.
Published: (2008) -
Protecting cyber physical systems using neural networks
by: Koshy, Ajay Philip
Published: (2022)