SwitchNet: a neural network model for forward and inverse scattering problems

We propose a novel neural network architecture, SwitchNet, for solving the wave equation based inverse scattering problems via providing maps between the scatterers and the scattered field (and vice versa). The main difficulty of using a neural network for this problem is that a scatterer has a global impact on the scattered wave field, rendering typical convolutional neural network with local connections inapplicable. While it is possible to deal with such a problem using a fully connected network, the number of parameters grows quadratically with the size of the input and output data. By leveraging the inherent low-rank structure of the scattering problems and introducing a novel switching layer with sparse connections, the SwitchNet architecture uses much fewer parameters and facilitates the training process. Numerical experiments show promising accuracy in learning the forward and inverse maps between the scatterers and the scattered wave field.

[1]  Lexing Ying,et al.  A Multiscale Neural Network Based on Hierarchical Matrices , 2018, Multiscale Model. Simul..

[2]  Leslie Greengard,et al.  High Resolution Inverse Scattering in Two Dimensions Using Recursive Linearization , 2016, SIAM J. Imaging Sci..

[3]  Jean-Pierre Berenger,et al.  A perfectly matched layer for the absorption of electromagnetic waves , 1994 .

[4]  Dimitrios I. Fotiadis,et al.  Artificial neural networks for solving ordinary and partial differential equations , 1997, IEEE Trans. Neural Networks.

[5]  R B Paris Hadamard 코드를 이용한 음성인식 무선덤웨이터의 구현 , 2011 .

[6]  Yingzhou Li,et al.  Butterfly-Net: Optimal Function Representation Based on Convolutional Neural Networks , 2018, Communications in Computational Physics.

[7]  Lexing Ying,et al.  Sparse Fourier Transform via Butterfly Algorithm , 2008, SIAM J. Sci. Comput..

[8]  P. Cochat,et al.  Et al , 2008, Archives de pediatrie : organe officiel de la Societe francaise de pediatrie.

[9]  Razvan Pascanu,et al.  On the difficulty of training recurrent neural networks , 2012, ICML.

[10]  Lexing Ying,et al.  Solving for high-dimensional committor functions using artificial neural networks , 2018, Research in the Mathematical Sciences.

[11]  Guigang Zhang,et al.  Deep Learning , 2016, Int. J. Semantic Comput..

[12]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[13]  Jürgen Schmidhuber,et al.  Deep learning in neural networks: An overview , 2014, Neural Networks.

[14]  Laurent Demanet,et al.  A Fast Butterfly Algorithm for the Computation of Fourier Integral Operators , 2008, Multiscale Model. Simul..

[15]  Tsuyoshi Murata,et al.  {m , 1934, ACML.

[16]  Yingzhou Li,et al.  Interpolative Butterfly Factorization , 2016, SIAM J. Sci. Comput..

[17]  Xudong Chen,et al.  Deep-Learning Schemes for Full-Wave Nonlinear Inverse Scattering Problems , 2019, IEEE Transactions on Geoscience and Remote Sensing.

[18]  Armin Lechleiter,et al.  MUSIC for Extended Scatterers as an Instance of the Factorization Method , 2009, SIAM J. Appl. Math..

[19]  E Weinan,et al.  The Deep Ritz Method: A Deep Learning-Based Numerical Algorithm for Solving Variational Problems , 2017, Communications in Mathematics and Statistics.

[20]  Geoffrey E. Hinton,et al.  Reducing the Dimensionality of Data with Neural Networks , 2006, Science.

[21]  Thomas Brox,et al.  U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.

[22]  Junshan Lin,et al.  Inverse scattering problems with multi-frequencies , 2015 .

[23]  Arnulf Jentzen,et al.  Solving high-dimensional partial differential equations using deep learning , 2017, Proceedings of the National Academy of Sciences.

[24]  Hongkai Zhao,et al.  Approximate Separability of the Green's Function of the Helmholtz Equation in the High Frequency Limit , 2018 .

[25]  E. Weinan,et al.  Deep Potential: a general representation of a many-body potential energy surface , 2017, 1707.01478.

[26]  Geoffrey E. Hinton,et al.  Deep Learning , 2015, Nature.

[27]  F. Natterer The Mathematics of Computerized Tomography , 1986 .

[28]  Bin Dong,et al.  PDE-Net: Learning PDEs from Data , 2017, ICML.

[29]  Lexing Ying,et al.  Butterfly Factorization , 2015, Multiscale Model. Simul..

[30]  Silvia Ferrari,et al.  A constrained integration (CINT) approach to solving partial differential equations using artificial neural networks , 2015, Neurocomputing.

[31]  Stig Larsson,et al.  Partial differential equations with numerical methods , 2003, Texts in applied mathematics.

[32]  Matthias Troyer,et al.  Solving the quantum many-body problem with artificial neural networks , 2016, Science.

[33]  R. Kress,et al.  Inverse Acoustic and Electromagnetic Scattering Theory , 1992 .

[34]  Lexing Ying,et al.  Solving parametric PDE problems with artificial neural networks , 2017, European Journal of Applied Mathematics.

[35]  Lukás Burget,et al.  Recurrent neural network based language model , 2010, INTERSPEECH.