DeepAI AI Chat
Log In Sign Up

A Phase Shift Deep Neural Network for High Frequency Wave Equations in Inhomogeneous Media

by   Wei Cai, et al.

In this paper, we propose a phase shift deep neural network (PhaseDNN) which provides a wideband convergence in approximating high frequency solutions of wave equations. The PhaseDNN accounts for the fact that many DNN achieves convergence in the low frequency range first, a series of moderately-sized of DNNs are constructed and trained for selected high frequency ranges. With the help of phase shifts in the frequency domain, each DNN will be trained to approximate the target solution's higher frequency content over a specific range. Due to the phase shift, each DNN achieves the speed of convergence as in the low frequency range. As a result, the proposed PhaseDNN is able to convert high frequency learning to low frequency learning, thus allowing a uniform learning to wideband high frequency functions. The PhaseDNN will then be applied to find the solution of high frequency wave equations in inhomogeneous media. Numerical results have demonstrated the capability of PhaseDNN in learning high frequency functions and oscillatory solutions of Helmholtz equations.


page 16

page 17

page 18

page 19

page 20

page 21

page 22

page 23


PhaseDNN - A Parallel Phase Shift Deep Neural Network for Adaptive Wideband Learning

In this paper, we propose a phase shift deep neural network (PhaseDNN) w...

An Upper Limit of Decaying Rate with Respect to Frequency in Deep Neural Network

Deep neural network (DNN) usually learns the target function from low to...

Sparse Approximate Multifrontal Factorization with Butterfly Compression for High Frequency Wave Equations

We present a fast and approximate multifrontal solver for large-scale sp...

On the Convergence of the Multi-scale Deep Neural Network (MscaleDNN) in Approximating Oscillatory Functions

In this paper, we derive diffusion models for the error evolution for a ...

PINNup: Robust neural network wavefield solutions using frequency upscaling and neuron splitting

Solving for the frequency-domain scattered wavefield via physics-informe...

Recovering Geometric Information with Learned Texture Perturbations

Regularization is used to avoid overfitting when training a neural netwo...