Conditional Wasserstein generative adversarial networks applied to acoustic metamaterial design

被引:16
|
作者
Lai, Peter [1 ]
Amirkulova, Feruza [1 ]
Gerstoft, Peter [2 ]
机构
[1] San Jose State Univ, Mech Engn Dept, San Jose, CA 95192 USA
[2] Univ Calif San Diego, Marine Phys Lab, Scripps Inst Oceanog, San Diego, CA 92037 USA
来源
关键词
INVERSE DESIGN; DEEP; AUTOENCODER; SCATTERING;
D O I
10.1121/10.0008929
中图分类号
O42 [声学];
学科分类号
070206 ; 082403 ;
摘要
This work presents a method for the reduction of the total scattering cross section (TSCS) for a planar configuration of cylinders by means of generative modeling and deep learning. Currently, the minimization of TSCS requires repeated forward modelling at considerable computer resources, whereas deep learning can do this more efficiently. The conditional Wasserstein generative adversarial networks (cWGANs) model is proposed for minimization of TSCS in two dimensions by combining Wasserstein generative adversarial networks with convolutional neural networks to simulate TSCS of configuration of rigid scatterers. The proposed cWGAN model is enhanced by adding to it a coordinate convolution (CoordConv) layer. For a given number of cylinders, the cWGAN model generates images of 2D configurations of cylinders that minimize the TSCS. The proposed generative model is illustrated with examples for planar uniform configurations of rigid cylinders.
引用
收藏
页码:4362 / 4374
页数:13
相关论文
共 50 条
  • [1] Wasserstein Generative Adversarial Networks
    Arjovsky, Martin
    Chintala, Soumith
    Bottou, Leon
    [J]. INTERNATIONAL CONFERENCE ON MACHINE LEARNING, VOL 70, 2017, 70
  • [2] Stable parallel training of Wasserstein conditional generative adversarial neural networks
    Lupo Pasini, Massimiliano
    Yin, Junqi
    [J]. JOURNAL OF SUPERCOMPUTING, 2023, 79 (02): : 1856 - 1876
  • [3] Conditional Wasserstein Generative Adversarial Networks for Rebalancing Iris Image Datasets
    Li, Yung-Hui
    Aslam, Muhammad Saqlain
    Harfiya, Latifa Nabila
    Chang, Ching-Chun
    [J]. IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS, 2021, E104D (09) : 1450 - 1458
  • [4] Stable Parallel Training of Wasserstein Conditional Generative Adversarial Neural Networks
    Pasini, Massimiliano Lupo
    Yin, Junqi
    [J]. 2021 INTERNATIONAL CONFERENCE ON COMPUTATIONAL SCIENCE AND COMPUTATIONAL INTELLIGENCE (CSCI 2021), 2021, : 1 - 7
  • [5] Stable parallel training of Wasserstein conditional generative adversarial neural networks
    Massimiliano Lupo Pasini
    Junqi Yin
    [J]. The Journal of Supercomputing, 2023, 79 : 1856 - 1876
  • [6] CONDITIONAL GENERATIVE ADVERSARIAL NETWORKS FOR ACOUSTIC ECHO CANCELLATION
    Pastor-Naranjo, Fran
    del Amor, Rocio
    Silva-Rodriguez, Julio
    Ferrer, Miguel
    Pinero, Gema
    Naranjo, Valery
    [J]. 2022 30TH EUROPEAN SIGNAL PROCESSING CONFERENCE (EUSIPCO 2022), 2022, : 85 - 89
  • [7] TOPOLOGY DESIGN WITH CONDITIONAL GENERATIVE ADVERSARIAL NETWORKS
    Sharpe, Conner
    Seepersad, Carolyn Conner
    [J]. PROCEEDINGS OF THE ASME INTERNATIONAL DESIGN ENGINEERING TECHNICAL CONFERENCES AND COMPUTERS AND INFORMATION IN ENGINEERING CONFERENCE, 2019, VOL 2A, 2020,
  • [8] De Novo Protein Design for Novel Folds Using Guided Conditional Wasserstein Generative Adversarial Networks
    Karimi, Mostafa
    Zhu, Shaowen
    Cao, Yue
    Shen, Yang
    [J]. JOURNAL OF CHEMICAL INFORMATION AND MODELING, 2020, 60 (12) : 5667 - 5681
  • [9] Inferential Wasserstein generative adversarial networks
    Chen, Yao
    Gao, Qingyi
    Wang, Xiao
    [J]. JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-STATISTICAL METHODOLOGY, 2022, 84 (01) : 83 - 113
  • [10] Single Image Haze Removal Using Conditional Wasserstein Generative Adversarial Networks
    Ebenezer, Joshua Peter
    Das, Bijaylaxmi
    Mukhopadhyay, Sudipta
    [J]. 2019 27TH EUROPEAN SIGNAL PROCESSING CONFERENCE (EUSIPCO), 2019,