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

2017 Fiscal Year Annual Research Report

Deepening and applications of sparse modeling by approaches of semiparametric Bayesian inference

Planned Research

Project AreaInitiative for High-Dimensional Data-Driven Science through Deepening of Sparse Modeling
Project/Area Number 25120012
Research InstitutionThe Institute of Statistical Mathematics

Principal Investigator

福水 健次  統計数理研究所, 数理・推論研究系, 教授 (60311362)

Co-Investigator(Kenkyū-buntansha) 鈴木 大慈  東京大学, 大学院情報理工学系研究科, 准教授 (60551372)
西山 悠  電気通信大学, 大学院情報理工学研究科, 助教 (60586395)
Project Period (FY) 2013-06-28 – 2018-03-31
Keywordsスパースモデリング / セミパラメトリック / ベイズ推論 / 最適化 / アルゴリズム
Outline of Annual Research Achievements

【課題1】「カーネルベイズ推論」に関しては,カーネル法によるStein作用素を利用したノンパラメトリックな適合度検定に関する研究を行い,高速に分布の適合度合と不適合な領域を同定する方法を提案した.この結果をまとめた論文は,機械学習分野の最難関国際会議であるNeural Information Processing SystemsにおいてBest Paper Awardを受賞した.また,シミュレータなどにより生成されたデータを用いた推論問題など,尤度計算が陽にできない場合のパラメータ推定の方法に関して研究を行い,有効は方法を提案した.その成果は機械学習分野のトップ国際会議のひとつ ICML 2018に採択された.
【課題2】「セミパラメトリック状態空間モデル」に関しては,モデルベースのカーネルベイズ則によるスムージング法に関する投稿論文の改訂を行い,審査結果を待っている段階である.また,尤度計算不能な場合の状態空間モデルのカーネルベイズアルゴリズムに関して海外共同研究者とともに研究を進め,論文を準備中である.
【課題3】「セミパラメトリックスパースモデリング」に関しては,特に深層学習や敵対的生成ネットワークモデルとその学習に関して,海外の共同研究者の含めて研究を推進し,NIPS, ICML, AISTATSなど機械学習分野のトップ国際会議に多数の論文を発表した.
【課題4】「スパース・セミパラメトリック表現」に関しては,微分方程式とスパースモデリングを共に用いた状態空間モデルの推定法に関して研究を行い,有効な方法を考案した.論文を準備中である.

Research Progress Status

29年度が最終年度であるため、記入しない。

Strategy for Future Research Activity

29年度が最終年度であるため、記入しない。

  • Research Products

    (21 results)

All 2018 2017 Other

All Int'l Joint Research (5 results) Journal Article (9 results) (of which Int'l Joint Research: 2 results,  Peer Reviewed: 9 results,  Open Access: 8 results) Presentation (5 results) (of which Int'l Joint Research: 3 results,  Invited: 5 results) Remarks (1 results) Funded Workshop (1 results)

  • [Int'l Joint Research] University College London/Oxford University/University of Bristol(英国)

    • Country Name
      UNITED KINGDOM
    • Counterpart Institution
      University College London/Oxford University/University of Bristol
  • [Int'l Joint Research] Max Planck Institute(ドイツ)

    • Country Name
      GERMANY
    • Counterpart Institution
      Max Planck Institute
  • [Int'l Joint Research] Mahidol University(タイ)

    • Country Name
      THAILAND
    • Counterpart Institution
      Mahidol University
  • [Int'l Joint Research] Pennsylvania State University(米国)

    • Country Name
      U.S.A.
    • Counterpart Institution
      Pennsylvania State University
  • [Int'l Joint Research] Universite Paris 6(フランス)

    • Country Name
      FRANCE
    • Counterpart Institution
      Universite Paris 6
  • [Journal Article] Generalized ridge estimator and model selection criteria in multivariate linear regression2018

    • Author(s)
      Mori Yuichi、Suzuki Taiji
    • Journal Title

      Journal of Multivariate Analysis

      Volume: 165 Pages: 243~261

    • DOI

      10.1016/j.jmva.2017.12.006

    • Peer Reviewed
  • [Journal Article] Independently Interpretable Lasso: A New Regularizer for Sparse Regression with Uncorrelated Variables2018

    • Author(s)
      M. Takada, T. Suzuki, H. Fujisawa
    • Journal Title

      Proceedings of the Twenty-First International Conference on Artificial Intelligence and Statistics, in PMLR

      Volume: 84 Pages: 454-463

    • Peer Reviewed / Open Access
  • [Journal Article] Gradient Layer: Enhancing the Convergence of Adversarial Training for Generative Models2018

    • Author(s)
      A. Nitanda, T. Suzuki
    • Journal Title

      Proceedings of the Twenty-First International Conference on Artificial Intelligence and Statistics, in PMLR

      Volume: 84 Pages: 1008-1016

    • Peer Reviewed / Open Access
  • [Journal Article] Fast generalization error bound of deep learning from a kernel perspective2018

    • Author(s)
      T. Suzuki
    • Journal Title

      Proceedings of the Twenty-First International Conference on Artificial Intelligence and Statistics, in PMLR

      Volume: 84 Pages: 1397-1406

    • Peer Reviewed / Open Access
  • [Journal Article] A Linear-Time Kernel Goodness-of-Fit Test2017

    • Author(s)
      Jitkrittum, W., Xu, W., Szabo, Z., Fukumizu, K. and Gretton, A.
    • Journal Title

      Advances in Neural Information Processing Systems (NIPS 2017)

      Volume: 30 Pages: 262--271

    • Peer Reviewed / Open Access / Int'l Joint Research
  • [Journal Article] Trimmed Density Ratio Estimation2017

    • Author(s)
      Liu, S., Takeda, A., Suzuki, T. and Fukumizu, K.
    • Journal Title

      Advances in Neural Information Processing Systems (NIPS 2017)

      Volume: 30 Pages: 4518--4528

    • Peer Reviewed / Open Access
  • [Journal Article] Doubly Accelerated Stochastic Variance Reduced Dual Averaging Method for Regularized Empirical Risk Minimization2017

    • Author(s)
      T. Murata, T. Suzuki
    • Journal Title

      Advances in Neural Information Processing Systems (NIPS2017)

      Volume: 30 Pages: 608--617

    • Peer Reviewed / Open Access
  • [Journal Article] Kernel Mean Embedding of Distributions: A Review and Beyond2017

    • Author(s)
      Muandet Krikamol、Fukumizu Kenji、Sriperumbudur Bharath、Sch?lkopf Bernhard
    • Journal Title

      Foundations and Trends? in Machine Learning

      Volume: 10 Pages: 1-141

    • DOI

      10.1561/2200000060

    • Peer Reviewed / Open Access / Int'l Joint Research
  • [Journal Article] Post Selection Inference with Kernels2017

    • Author(s)
      M. Yamada, Y. Umezu, K. Fukumizu, I. Takeuchi
    • Journal Title

      Proceedings of the Twenty-First International Conference on Artificial Intelligence and Statistics, in PMLR

      Volume: 84 Pages: 152-160

    • Peer Reviewed / Open Access
  • [Presentation] Generalization Error and Compressibility of Deep Learning via Kernel Analysis2018

    • Author(s)
      Taiji Suzuki
    • Organizer
      Tokyo Deep Learning Workshop (Deep Learning: Theory, Algorithms, and Applications)
    • Int'l Joint Research / Invited
  • [Presentation] Taxonomy Matching between Asteroids and Meteorites: Supervised Clustering Approach2017

    • Author(s)
      Fukumizu, K., Saito, Y., Miyamoto, H., Niihara, T. and Peng, H.
    • Organizer
      High-Dimensional Data-Driven Science (HD3-2017)
    • Int'l Joint Research / Invited
  • [Presentation] 機械学習技術の進展とその数理基盤2017

    • Author(s)
      鈴木大慈
    • Organizer
      数理システムユーザーコンファレンス
    • Invited
  • [Presentation] Generalization error analysis of deep learning and its application to network structure determination2017

    • Author(s)
      Taiji Suzuki
    • Organizer
      French-Japanese Workshop on Deep Learning and Artificial Intelligence
    • Int'l Joint Research / Invited
  • [Presentation] 構造のある機械学習問題における最適化技法2017

    • Author(s)
      鈴木大慈
    • Organizer
      第29回RAMPシンポジウム, 企画セッション「機械学習と最適化」
    • Invited
  • [Remarks]

    • URL

      http://www.ism.ac.jp/~fukumizu/

  • [Funded Workshop] Workshop on Functional Inference and Machine Intelligence (FIMI)2018

URL: 

Published: 2018-12-17   Modified: 2022-03-29  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi