• Search Research Projects
  • Search Researchers
  • How to Use
  1. Back to previous page

Deep Quasi-Linear SVM Based on Deep Neural Network and Its Applications

Research Project

Project/Area Number 17K06506
Research Category

Grant-in-Aid for Scientific Research (C)

Allocation TypeMulti-year Fund
Section一般
Research Field Control engineering/System engineering
Research InstitutionWaseda University

Principal Investigator

HU Jinglu  早稲田大学, 理工学術院(情報生産システム研究科・センター), 教授 (50294905)

Project Period (FY) 2017-04-01 – 2020-03-31
Project Status Completed (Fiscal Year 2019)
Budget Amount *help
¥4,680,000 (Direct Cost: ¥3,600,000、Indirect Cost: ¥1,080,000)
Fiscal Year 2019: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
Fiscal Year 2018: ¥1,950,000 (Direct Cost: ¥1,500,000、Indirect Cost: ¥450,000)
Fiscal Year 2017: ¥1,690,000 (Direct Cost: ¥1,300,000、Indirect Cost: ¥390,000)
Keywords深層ニューラルネットワーク / サポートベクターマシン / 深層学習 / 機械学習 / パターン認識
Outline of Final Research Achievements

In this research, a quasi-linear support vector machine (SVM) is constructed by using deep neural networks, in which a deep quasi-linear kernel is composed by using deep learning. By using the deep quasi-linear SVM, it is possible to realize a deep learning in the cases where there is rather few data. First, pretrain a deep neural network via transfer learning for composing a deep quasi-linear kernel; Then an SVM with the quasi-linear kernel can be obtained using the few data.

Academic Significance and Societal Importance of the Research Achievements

本研究では、サポートベクターマシン(SVM)を深層学習で訓練済みの深層ニューラルネットワークから構築する。SVMのための深層準線形カーネルの構築に通してSVMと深層ニューラルネットワークとの間に橋を架け、近年著しく発展できている深層学習技術を活用し、深層カーネルの学習は大規模なデータの場合でも容易に実現できる。一方、深層学習の立場から見れば、訓練済の深層ニューラルネットワークからSVMのカーネルを合成し(転移学習)、このカーネルに基づいた分類器をSVM 最適化(小データ)することによって、小データでも深層学習の実現が可能になる。

Report

(4 results)
  • 2019 Annual Research Report   Final Research Report ( PDF )
  • 2018 Research-status Report
  • 2017 Research-status Report
  • Research Products

    (20 results)

All 2020 2019 2018 2017

All Journal Article (20 results) (of which Peer Reviewed: 20 results)

  • [Journal Article] Fast SVM Training using Data Reconstruction for Classification of Very Large Datasets2020

    • Author(s)
      P. Liang, W. Li and J. Hu
    • Journal Title

      IEEJ Trans. on Electrical and Electronic Engineering C

      Volume: 15(3) Issue: 3 Pages: 372-381

    • DOI

      10.1002/tee.23065

    • Related Report
      2019 Annual Research Report
    • Peer Reviewed
  • [Journal Article] An Autoencoder Based Piecewise Linear Model for Nonlinear Classification using Quasi-Linear Support Vector Machines2019

    • Author(s)
      W. Li, P. Liang and J. Hu
    • Journal Title

      IEEJ Trans. on Electrical and Electronic Engineering C

      Volume: 14(8) Issue: 8 Pages: 1236-1243

    • DOI

      10.1002/tee.22923

    • Related Report
      2019 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Air Quality Forecasting using SVR with Quasi-Linear Kernel2019

    • Author(s)
      H. Zhu and J. Hu
    • Journal Title

      Proc. of the 2019 International Conference on Computer, Information and Telecommunication Systems (CITS'2019) (Bejing)

      Volume: 2019(8) Pages: 126-130

    • Related Report
      2019 Annual Research Report
    • Peer Reviewed
  • [Journal Article] A Deep Neural Network Based Hierarchical Multi-LabelClassifier for Protein Function Prediction2019

    • Author(s)
      X. Yuan, W. Li, K. Lin and J. Hu
    • Journal Title

      Proc. of the 2019 International Conference on Computer, Information and Telecommunication Systems (CITS'2019) (Bejing)

      Volume: 2019(8) Pages: 131-135

    • Related Report
      2019 Annual Research Report
    • Peer Reviewed
  • [Journal Article] A Semi-Supervised Classification Using Gated Linear Model2019

    • Author(s)
      Y. Ren, W. Li and J. Hu
    • Journal Title

      Proc. of 2017 IEEE International Joint Conference on Neural Networks (IJCNN'2019) (Budapest)

      Volume: 2019(7) Pages: 1-7

    • Related Report
      2019 Annual Research Report
    • Peer Reviewed
  • [Journal Article] A coarse‐to‐fine two‐step method for semisupervised classification using quasi‐linear Laplacian SVM2019

    • Author(s)
      Zhou Bo、Li Weite、Hu Jinglu
    • Journal Title

      IEEJ Transactions on Electrical and Electronic Engineering

      Volume: 14(3) Issue: 3 Pages: 441-448

    • DOI

      10.1002/tee.22825

    • Related Report
      2018 Research-status Report
    • Peer Reviewed
  • [Journal Article] One-Class Classification Using Support Vector Machine with Quasi-Linear Kernel2019

    • Author(s)
      P. Liang, W. Li, H. Tian and J. Hu
    • Journal Title

      IEEJ Trans. on Electrical and Electronic Engineering C

      Volume: 14(3) Issue: 3 Pages: 449-456

    • DOI

      10.1002/tee.22826

    • Related Report
      2018 Research-status Report
    • Peer Reviewed
  • [Journal Article] Quasi-Linear SVM Classifier with Segmented Local Offsets for Imbalanced Data Classification2019

    • Author(s)
      P. Liang, F. Zheng, W. Li and J. Hu
    • Journal Title

      IEEJ Trans. on Electrical and Electronic Engineering C

      Volume: 14(2) Issue: 2 Pages: 289-296

    • DOI

      10.1002/tee.22808

    • Related Report
      2018 Research-status Report
    • Peer Reviewed
  • [Journal Article] Oversampling the Minority Class in a Multi-Linear Feature Space for Imbalanced Data Classification2018

    • Author(s)
      P. Liang, W. Li and J. Hu
    • Journal Title

      IEEJ Trans. on Electrical and Electronic Engineering C

      Volume: 13(10) Issue: 10 Pages: 1483-1491

    • DOI

      10.1002/tee.22715

    • Related Report
      2018 Research-status Report
    • Peer Reviewed
  • [Journal Article] A Metric Learning Method for Improving Neural Network Based Kernel Learning for SVM2018

    • Author(s)
      P. Liang, X. Yao and J. Hu
    • Journal Title

      Proc. of IEEE Inter. Conference on Systems, Man and Cybernetics (SMC'2018) (Miyazaki)

      Volume: 2018(10) Pages: 1637-1642

    • Related Report
      2018 Research-status Report
    • Peer Reviewed
  • [Journal Article] A Convolutional AutoEncoder Method for Anomaly Detection on System Logs2018

    • Author(s)
      Y. Cui, Y.P. Sun and J. Hu
    • Journal Title

      Proc. of IEEE Inter. Conference on Systems, Man and Cybernetics (SMC'2018) (Miyazaki)

      Volume: 2018(10) Pages: 3053-3058

    • Related Report
      2018 Research-status Report
    • Peer Reviewed
  • [Journal Article] Feature Extraction using a Mutually-Competitive Autoencoder for Protein Function Prediction2018

    • Author(s)
      L.J.V. Miranda and J. Hu
    • Journal Title

      Proc. of IEEE Inter. Conference on Systems, Man and Cybernetics (SMC'2018) (Miyazaki)

      Volume: 2018(10) Pages: 1333-1338

    • Related Report
      2018 Research-status Report
    • Peer Reviewed
  • [Journal Article] A Deep Learning Approach Based on Stacked Denoising Autoencoders for Protein Function Prediction2018

    • Author(s)
      L.J.V. Miranda and J. Hu
    • Journal Title

      Proc. of the 42th IEEE Inter. Conference on Computers, Software & Applications (COMPSAC'2018) (Tokyo)

      Volume: 2018(7) Pages: 480-485

    • Related Report
      2018 Research-status Report
    • Peer Reviewed
  • [Journal Article] A Geometry-Based Two-step Method for Nonlinear Classification Using Quasi-Linear Support Vector Machine2017

    • Author(s)
      W.Li, B.Zhou, B.Chen and J.Hu
    • Journal Title

      IEEJ Trans. on Electrical and Electronic Engineering

      Volume: 12(6) Pages: 883-889

    • Related Report
      2017 Research-status Report
    • Peer Reviewed
  • [Journal Article] A New Segmented Oversampling Method for Imbalanced Data Classification Using Quasi-Linear Support Vector Machine2017

    • Author(s)
      B.Zhou, W.Li and J.Hu
    • Journal Title

      IEEJ Trans. on Electrical and Electronic Engineering

      Volume: 12(6) Pages: 891-898

    • Related Report
      2017 Research-status Report
    • Peer Reviewed
  • [Journal Article] Stacked Residual Recurrent Neural Network with Word Weight for Text Classification2017

    • Author(s)
      W.Cao, A.Song and J.Hu
    • Journal Title

      IAENG International Journal of Computer Science

      Volume: 44(3) Pages: 277-284

    • Related Report
      2017 Research-status Report
    • Peer Reviewed
  • [Journal Article] Large-Scale Image Classification Using Fast SVM with Deep Quasi-Linear Kernel2017

    • Author(s)
      P. Liang, W. Li, D. Liu and J. Hu
    • Journal Title

      Proc. of 2017 IEEE International Joint Conference on Neural Networks (IJCNN'2017) (Anchorage)

      Volume: 2017(5) Pages: 1064-1071

    • Related Report
      2017 Research-status Report
    • Peer Reviewed
  • [Journal Article] A Mixture of Multiple Linear Classifiers with Sample Weight and Manifold Regularization2017

    • Author(s)
      W. Li, B. Chen, B. Zhou and J. Hu
    • Journal Title

      Proc. of 2017 IEEE International Joint Conference on Neural Networks (IJCNN'2017) (Anchorage)

      Volume: 2017(5) Pages: 3747-3752

    • Related Report
      2017 Research-status Report
    • Peer Reviewed
  • [Journal Article] A Multilayer Gated Bilinear Classifier: from Optimizing a Deep Rectified Network to a Support Vector Machine2017

    • Author(s)
      W. Li and J. Hu
    • Journal Title

      Proc. of 2017 IEEE International Joint Conference on Neural Networks (IJCNN'2017) (Anchorage)

      Volume: 2017(5) Pages: 140-146

    • Related Report
      2017 Research-status Report
    • Peer Reviewed
  • [Journal Article] Non-Local Information for a Mixture of Multiple Linear Classifiers2017

    • Author(s)
      W. Li, P. Liang, X. Yuan and J. Hu
    • Journal Title

      Proc. of 2017 IEEE International Joint Conference on Neural Networks (IJCNN'2017) (Anchorage)

      Volume: 2017(5) Pages: 3741-3746

    • Related Report
      2017 Research-status Report
    • Peer Reviewed

URL: 

Published: 2017-04-28   Modified: 2021-02-19  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi