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

Analysis and Developments of Kernel Adaptive Filtering Algorithm

Research Project

Project/Area Number 15K06081
Research Category

Grant-in-Aid for Scientific Research (C)

Allocation TypeMulti-year Fund
Section一般
Research Field Communication/Network engineering
Research InstitutionKeio University

Principal Investigator

Yukawa Masahiro  慶應義塾大学, 理工学部(矢上), 准教授 (60462743)

Project Period (FY) 2015-04-01 – 2019-03-31
Project Status Completed (Fiscal Year 2018)
Budget Amount *help
¥4,940,000 (Direct Cost: ¥3,800,000、Indirect Cost: ¥1,140,000)
Fiscal Year 2017: ¥1,560,000 (Direct Cost: ¥1,200,000、Indirect Cost: ¥360,000)
Fiscal Year 2016: ¥1,690,000 (Direct Cost: ¥1,300,000、Indirect Cost: ¥390,000)
Fiscal Year 2015: ¥1,690,000 (Direct Cost: ¥1,300,000、Indirect Cost: ¥390,000)
Keywords再生核ヒルベルト空間 / 直積空間 / オンライン学習 / カーネル法 / 非線形関数推定 / 凸射影 / 時系列データ予測 / 適応フィルタ / 再生核 / 非線形推定 / 直交射影 / 適応学習
Outline of Final Research Achievements

Nonlinearity is encountered in many situations in signal processing and data science, such as in characteristics of low-cost speaker, channel of optical communication, relationship between the past and future measurements of time-series data, to name a few. Kernel adaptive filtering has good tradeoffs among estimation accuracy, complexity, and global optimality (convexity).
In the current research project, analysis and performance improvements of multikernel adaptive filtering have been addressed. The multikernel adaptive filtering is insensitive to the design of kernels and is able to capture the multiscaleness of data. The outcomes of the project include the developments of fast converging algorithms with understanding of its mechanism, an extension to Hilbert spaces that have no reproducing kernels, and an extension to distributed settings with sensor networks envisioned.

Academic Significance and Societal Importance of the Research Achievements

モデル選択問題は、数理科学における重要テーマとして長年研究されてきた。多カーネル適応フィルタは、複数の再生核を利用することで柔軟で冗長な数理モデルを用意し、スパース化によって冗長な辞書から適切な要素のみを自動抽出する機能を備えた手法である。研究代表者は、同手法に関する研究に2010年に着手し、2012年に誌上発表、以降、様々な方向に研究を進展させてきた。本科研費プロジェクトで得られた成果は、アルゴリズムの高速化・高速化メカニズムの理解・分散型アルズリズムへの発展を含むものであり、ビッグデータ解析への需要と期待が高まる現代と未来において、基盤技術としての役割を担っていくことを期待する。

Report

(5 results)
  • 2018 Annual Research Report   Final Research Report ( PDF )
  • 2017 Research-status Report
  • 2016 Research-status Report
  • 2015 Research-status Report
  • Research Products

    (28 results)

All 2018 2017 2016 2015 Other

All Int'l Joint Research (5 results) Journal Article (6 results) (of which Int'l Joint Research: 2 results,  Peer Reviewed: 6 results,  Acknowledgement Compliant: 3 results) Presentation (15 results) (of which Int'l Joint Research: 14 results) Remarks (2 results)

  • [Int'l Joint Research] Technical University of Berlin/Fraunhofer HHI/University of Bremen(ドイツ)

    • Related Report
      2018 Annual Research Report
  • [Int'l Joint Research] ブレーメン大学(ドイツ)

    • Related Report
      2017 Research-status Report
  • [Int'l Joint Research] Fraunhofer Heinrich Hertz Institute(ドイツ)

    • Related Report
      2017 Research-status Report
  • [Int'l Joint Research] ベルリン大学(ドイツ)

    • Related Report
      2017 Research-status Report
  • [Int'l Joint Research] ベルリン工科大学(ドイツ)

    • Related Report
      2016 Research-status Report
  • [Journal Article] Distributed Adaptive Learning With Multiple Kernels in Diffusion Networks2018

    • Author(s)
      Shin Ban-Sok、Yukawa Masahiro、Cavalcante Renato Luis Garrido、Dekorsy Armin
    • Journal Title

      IEEE Transactions on Signal Processing

      Volume: 66 Issue: 21 Pages: 5505-5519

    • DOI

      10.1109/tsp.2018.2868040

    • Related Report
      2018 Annual Research Report
    • Peer Reviewed / Int'l Joint Research
  • [Journal Article] Online Nonlinear Estimation via Iterative L2-Space Projections: Reproducing Kernel of Subspace2018

    • Author(s)
      Ohnishi Motoya、Yukawa Masahiro
    • Journal Title

      IEEE Transactions on Signal Processing

      Volume: 66 Issue: 15 Pages: 4050-4064

    • DOI

      10.1109/tsp.2018.2846271

    • Related Report
      2018 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Supervised Nonnegative Matrix Factorization via Minimization of Regularized Moreau-Envelope of Divergence Function with Application to Music Transcription2018

    • Author(s)
      Masahiro Yukawa and Hideaki Kagami
    • Journal Title

      Journal of the Franklin Institute

      Volume: 355 Issue: 4 Pages: 2041-2066

    • DOI

      10.1016/j.jfranklin.2017.12.002

    • Related Report
      2017 Research-status Report
    • Peer Reviewed
  • [Journal Article] Efficient dictionary-refining kernel adaptive filter with fundamental insights2016

    • Author(s)
      Masaaki Takizawa and Masahiro Yukawa
    • Journal Title

      IEEE Trans. Signal Processing

      Volume: 64 Issue: 16 Pages: 4337-4350

    • DOI

      10.1109/tsp.2016.2563401

    • Related Report
      2016 Research-status Report
    • Peer Reviewed / Acknowledgement Compliant
  • [Journal Article] Why does a Hilbertian metric work efficiently in online learning with kernels?2016

    • Author(s)
      Masahiro Yukawa and Klaus R. Muller
    • Journal Title

      IEEE Signal Processing Letters

      Volume: 23 Issue: 10 Pages: 1424-1428

    • DOI

      10.1109/lsp.2016.2598615

    • Related Report
      2016 Research-status Report
    • Peer Reviewed / Int'l Joint Research / Acknowledgement Compliant
  • [Journal Article] Adaptive learning in Cartesian product of reproducing kernel Hilbert spaces2015

    • Author(s)
      Masahiro Yukawa
    • Journal Title

      IEEE Trans. Signal Processing

      Volume: 63 Issue: 22 Pages: 6037-6048

    • DOI

      10.1109/tsp.2015.2463261

    • Related Report
      2015 Research-status Report
    • Peer Reviewed / Acknowledgement Compliant
  • [Presentation] Joint separation and dereverberation of reverberant mixtures with determined multichannel non-negative matrix factorization2018

    • Author(s)
      Hideaki Kagami, Hirokazu Kameoka, and Masahiro Yukawa
    • Organizer
      International Conference on Acoustic, Speech and Signal Processing
    • Related Report
      2018 Annual Research Report
    • Int'l Joint Research
  • [Presentation] A hybrid dictionary approach for distributed kernel adaptive filtering in diffusion networks2018

    • Author(s)
      Ban-Sok Shin, Masahiro Yukawa, Renato L.G. Cavalcante, and Armin Dekorsy
    • Organizer
      International Conference on Acoustic, Speech and Signal Processing
    • Related Report
      2018 Annual Research Report
    • Int'l Joint Research
  • [Presentation] How Are the Centered Kernel Principal Components Relevant to Regression Task?2018

    • Author(s)
      Masahiro Yukawa, Klaus R. Mueller, and Yuto Ogino
    • Organizer
      International Conference on Acoustic, Speech and Signal Processing
    • Related Report
      2018 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Automatic shrinkage tuning robust to input correlation for sparsity-aware adaptive filtering2018

    • Author(s)
      Kwangjin Jeong, Masahiro Yukawa, Masao Yamagishi, and Isao Yamada
    • Organizer
      International Conference on Acoustic, Speech and Signal Processing
    • Related Report
      2018 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Detection for 5G-NOMA: An Online Adaptive Machine Learning Approach2018

    • Author(s)
      Daniyal Amir Awan, Renato L.G. Cavalcante, Masahiro Yukawa, and Slawomir Stanczak
    • Organizer
      IEEE International Conference on Communications
    • Related Report
      2018 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Spectral Clustering with Automatic Cluster-Number Identification via Finding Sparse Eigenvectors2018

    • Author(s)
      Yuto Ogino and Masahiro Yukawa
    • Organizer
      European Signal Processing Conference
    • Related Report
      2018 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Automatic shrinkage tuning based on a system-mismatch estimate for sparsity-aware adaptive filtering2017

    • Author(s)
      Masao Yamagishi, Yukawa Masahiro, and Isao Yamada
    • Organizer
      IEEE ICASSP
    • Place of Presentation
      ニューオリンズ・米国
    • Year and Date
      2017-03-05
    • Related Report
      2016 Research-status Report
    • Int'l Joint Research
  • [Presentation] A majorization-minimization algorithm with projected gradient updates for time-domain spectrogram factorization2017

    • Author(s)
      Hideaki Kagami, Hirokazu Kameoka, and Masahiro Yukawa
    • Organizer
      IEEE ICASSP
    • Place of Presentation
      ニューオリンズ・米国
    • Year and Date
      2017-03-05
    • Related Report
      2016 Research-status Report
    • Int'l Joint Research
  • [Presentation] Projection-based dual averaging for stochastic sparse optimization2017

    • Author(s)
      Asahi Ushio and Masahiro Yukawa
    • Organizer
      IEEE ICASSP
    • Place of Presentation
      ニューオリンズ・米国
    • Year and Date
      2017-03-05
    • Related Report
      2016 Research-status Report
    • Int'l Joint Research
  • [Presentation] Complex NMF with the generalized Kullback-Leibler divergence2017

    • Author(s)
      Hirokazu Kameoka, Hideaki Kagami, and Masahiro Yukawa
    • Organizer
      IEEE ICASSP
    • Place of Presentation
      ニューオリンズ・米国
    • Year and Date
      2017-03-04
    • Related Report
      2016 Research-status Report
    • Int'l Joint Research
  • [Presentation] Distributed Nonlinear Regression Using In-Network Processing With Multiple Gaussian Kernels2017

    • Author(s)
      B.-S. Shin, H. Paul, M. Yukawa, and A. Dekorsy
    • Organizer
      IEEE International Workshop on Signal Processing Advances in Wireless Communications
    • Related Report
      2017 Research-status Report
    • Int'l Joint Research
  • [Presentation] Online Learning in L2 Space with Multiple Gaussian Kernels2017

    • Author(s)
      Motoya Ohnishi and Masahiro Yukawa
    • Organizer
      European Signal Processing Conference
    • Related Report
      2017 Research-status Report
    • Int'l Joint Research
  • [Presentation] A Better Metric in Kernel Adaptive Filtering2016

    • Author(s)
      Airi Takeuchi, Masahiro Yukawa, and Klaus R. Muller
    • Organizer
      EUSIPCO
    • Place of Presentation
      ブダペスト・ハンガリー
    • Year and Date
      2016-08-29
    • Related Report
      2016 Research-status Report
    • Int'l Joint Research
  • [Presentation] Supervised Nonnegative Matrix Factorization with Dual-Itakura-Saito and Kullback-Leibler Divergences for Music Transcription2016

    • Author(s)
      Hideaki Kagami and Masahiro Yukawa
    • Organizer
      EUSIPCO
    • Place of Presentation
      ブダペスト・ハンガリー
    • Year and Date
      2016-08-29
    • Related Report
      2016 Research-status Report
    • Int'l Joint Research
  • [Presentation] Multi-task Learning with Operator-valued Reproducing Kernels2015

    • Author(s)
      Asahi Ushio and Masahiro Yukawa
    • Organizer
      信号処理シンポジウム
    • Place of Presentation
      スパリゾートハワイアンズ(福島県・いわき市)
    • Year and Date
      2015-11-04
    • Related Report
      2015 Research-status Report
  • [Remarks] Masahiro Yukawa -- Publications

    • URL

      http://www.ykw.elec.keio.ac.jp/yukawa/pub.html

    • Related Report
      2017 Research-status Report 2015 Research-status Report
  • [Remarks] Masahiro Yukawa

    • URL

      http://www.ykw.elec.keio.ac.jp/yukawa/index.html

    • Related Report
      2016 Research-status Report

URL: 

Published: 2015-04-16   Modified: 2022-02-22  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi