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

2012 Fiscal Year Final Research Report

Fast Likelihood Ratio Optimization Based Upon Genaralized Logarithm and Its Applications

Research Project

  • PDF
Project/Area Number 22656088
Research Category

Grant-in-Aid for Challenging Exploratory Research

Allocation TypeSingle-year Grants
Research Field Communication/Network engineering
Research InstitutionWaseda University

Principal Investigator

MATSUYAMA Yasuo  早稲田大学, 理工学術院, 教授 (60125804)

Project Period (FY) 2010 – 2012
Keywords信号処理
Research Abstract

Likelihood optimization for learning algorithms was generalized by using the alpha-logarithm. This generalization led to a faster convergence than traditional methods. Algorithms on hidden Markov model estimation and independent component analysis were chosen since they have high ramifications. The use of the alpha-logarithm appears as the utilization of past information via momentum terms. This property enabled faster convergence than traditional methods.

  • Research Products

    (14 results)

All 2012 2011 2010 Other

All Journal Article (8 results) (of which Peer Reviewed: 7 results) Book (1 results) Remarks (3 results) Patent(Industrial Property Rights) (2 results)

  • [Journal Article] Rapid algorithm for independent component analysis2012

    • Author(s)
      R. Yokote and Y. Matsuyama
    • Journal Title

      Journal of Signal and Information Processing

      Volume: Vol.3, No.3 Pages: 275-285

    • DOI

      DOI:10.4236/jsip.2012.33037

    • Peer Reviewed
  • [Journal Article] Conversion of sensitivity-based tasks from brain signals and motions, Proceedings of the International Association of Science and Technology for Development's International2012

    • Author(s)
      Y. Matsuyama, R. Yokote and Y. Yokosawa
    • Journal Title

      Conference on Artificial Intelligence and Soft Computing

      Pages: 270-277

    • DOI

      DOI:10.2316/P.2012.777-015

    • Peer Reviewed
  • [Journal Article] From convex divergence to human-aware information processing : Good models mismatch well, therefore serviceable2012

    • Author(s)
      Y. Matsuyama
    • Journal Title

      Invited Talk Abstract of International Workshop on Anomalous Statistics, Generalized Entropies, and Information Geometry

      Pages: 20

    • URL

      https://sites.google.com/site/next2012nara/program

  • [Journal Article] Independent component analysis with graphical correlation : Applications to multi-vision coding, Proceedings of International Joint Conference on Neural Networks2011

    • Author(s)
      R. Yokote, T. Nakamura and Y. Matsuyama
    • Journal Title

      San Jose, CA

      Pages: 701-708

    • DOI

      DOI:10.1109/IJCNN.2011.6033290

    • Peer Reviewed
  • [Journal Article] Hidden Markov model estimation based on the alpha-EM algorithm : Discrete and continuous alpha-HMMs2011

    • Author(s)
      Y. Matsuyama
    • Journal Title

      Proceedings of International Joint Conference on Neural Networks

      Pages: 809-816

    • DOI

      DOI: 10.1109/IJCNN.2011.6033304

    • Peer Reviewed
  • [Journal Article] Fast estimation of hidden Markov models via alpha-EM algorithm2011

    • Author(s)
      Y. Matsuyama and R. Yokote
    • Journal Title

      Proceedings of 2011 IEEE Statistical Signal Processing Workshop

      Pages: 89-92

    • DOI

      DOI:10.1109/SSP.2011.5967835

    • Peer Reviewed
  • [Journal Article] Yet rapid ICA : Applications to un-indexed image-to-image retrieval2010

    • Author(s)
      R. Yokote and Y. Matsuyama
    • Journal Title

      Proceedings of International Joint Conference on Neural Networks

      Pages: 4255-4262

    • DOI

      DOI:10.1109/IJCNN.2010.5596895

    • Peer Reviewed
  • [Journal Article] Alpha-EM gives fast hidden Markov model estimation : Derivation and evaluation of alpha-HMM2010

    • Author(s)
      Y. Matsuyama and R. Hayashi
    • Journal Title

      Proceedings of International Joint Conference on Neural Networks

      Pages: 663-670

    • DOI

      DOI: 10.1109/IJCNN.2010.5596959

    • Peer Reviewed
  • [Book] バイオインフォマティクス in silico2011

    • Author(s)
      松山泰男
    • Total Pages
      290
    • Publisher
      培風館
  • [Remarks] 松山泰男、汎用性を有する高速な隠れマルコフモデル推定法、科学技術振興機構新技術説明会資料、科学技術振興機構JSTホール、2011年12月20日。

    • URL

      http://www.wiz.cs.waseda.ac.jp

  • [Remarks] 松山泰男研究室

    • URL

      http://researchmap.jp/read0169581/

  • [Remarks] ReaD & Researchmap

  • [Patent(Industrial Property Rights)] 隠れマルコフモデルの推定方法,推定装置および推定プログラム2010

    • Inventor(s)
      松山泰男, 林 龍之介
    • Industrial Property Rights Holder
      松山泰男, 林 龍之介と早稲田大学
    • Industrial Property Number
      特願2010-159949
    • Filing Date
      2010-07-16
  • [Patent(Industrial Property Rights)] 独立成分分析の分析装置,分析方法および分析プログラム2010

    • Inventor(s)
      松山泰男, 横手良太
    • Industrial Property Rights Holder
      松山泰男, 横手良太と早稲田大学
    • Industrial Property Number
      特願2010-161211
    • Filing Date
      2010-07-16

URL: 

Published: 2014-08-29  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi