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

Knowledge-embedded Bayesian deep learning and its application to small data analysis

Research Project

Project/Area Number 21H03511
Research Category

Grant-in-Aid for Scientific Research (B)

Allocation TypeSingle-year Grants
Section一般
Review Section Basic Section 61040:Soft computing-related
Research InstitutionOsaka University (2022-2023)
Kyushu University (2021)

Principal Investigator

Hayashi Hideaki  大阪大学, データビリティフロンティア機構, 准教授 (00790015)

Co-Investigator(Kenkyū-buntansha) 古居 彬  広島大学, 先進理工系科学研究科(工), 助教 (30868237)
Project Period (FY) 2021-04-01 – 2024-03-31
Project Status Completed (Fiscal Year 2023)
Budget Amount *help
¥17,420,000 (Direct Cost: ¥13,400,000、Indirect Cost: ¥4,020,000)
Fiscal Year 2023: ¥5,720,000 (Direct Cost: ¥4,400,000、Indirect Cost: ¥1,320,000)
Fiscal Year 2022: ¥5,200,000 (Direct Cost: ¥4,000,000、Indirect Cost: ¥1,200,000)
Fiscal Year 2021: ¥6,500,000 (Direct Cost: ¥5,000,000、Indirect Cost: ¥1,500,000)
Keywordsニューラルネットワーク / 深層学習 / ベイズ推定 / 希少データ / ベイズ推論
Outline of Research at the Start

深層学習は複数の分野で成功を収めてきたが,十分な精度を得るには大量の学習データが必要となる.そのため,医療データのようなデータ収集にコストがかかり,ラベリングに専門知識が必要なドメインでは適用が困難な場合が多い.そこで本研究では,希少データを適切に学習するための技術として知識埋め込み型ベイズニューラルネットワークを提案する.提案法では領域の知識を確率モデリングし,ネットワーク構造へ埋め込むことで少ないデータ数での学習を可能にする.また,ベイズ的に学習することによって,不確実性の推定や欠損値の補完など柔軟な確率計算を実現する.さらに,生体信号や医用画像といった実データ解析応用へ展開していく.

Outline of Final Research Achievements

Applying existing deep learning algorithms is often challenging for small data, such as medical data, where data collection is costly and annotation requires specialized knowledge. In this study, we proposed a framework for appropriately learning small data, called knowledge-embedded Bayesian deep learning. During the study, we developed the foundational techniques and applied them to real-world data analysis. Specifically, we proposed methods to reduce the amount of labeled data by estimating the distribution of input data while training the classifier, and methods to efficiently select training data through collaboration between the classifier and humans. Additionally, we performed real-world data analysis on various datasets such as infant motion images and electromyograms.

Academic Significance and Societal Importance of the Research Achievements

社会的意義としては,機械学習アルゴリズムを実データ解析に応用する際,データ収集やラベル付けに必要なコストを削減する技術を提案した点である.特に,医用データ解析では大規模データの収集そのものが難しく,ラベル付けにも医師の専門知識が必要となるため,有効な手段となる.学術的意義としては,単一の深層学習モデルで識別モデルと生成モデルを同時学習できる半教師あり学習や信頼度較正に有効な手法の提案や,相対ラベルを用いて学習できるベイズ深層学習モデルの提案とその妥当性の理論的証明が主な貢献となる.

Report

(4 results)
  • 2023 Annual Research Report   Final Research Report ( PDF )
  • 2022 Annual Research Report
  • 2021 Annual Research Report
  • Research Products

    (25 results)

All 2024 2023 2022 2021 Other

All Journal Article (5 results) (of which Peer Reviewed: 5 results,  Open Access: 2 results) Presentation (19 results) (of which Int'l Joint Research: 9 results) Remarks (1 results)

  • [Journal Article] A Hybrid of Generative and Discriminative Models Based on the Gaussian-Coupled Softmax Layer2024

    • Author(s)
      Hayashi Hideaki
    • Journal Title

      IEEE Transactions on Neural Networks and Learning Systems

      Volume: - Issue: 2 Pages: 2894-2904

    • DOI

      10.1109/tnnls.2024.3358113

    • Related Report
      2023 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Human-Machine Interfaces Based on Bioelectric Signals: A Narrative Review with a Novel System Proposal2022

    • Author(s)
      Hayashi Hideaki、Tsuji Toshio
    • Journal Title

      IEEJ Transactions on Electrical and Electronic Engineering

      Volume: 17 Issue: 11 Pages: 1536-1544

    • DOI

      10.1002/tee.23646

    • Related Report
      2022 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] A Neural Network Based on the Johnson SU Translation System and Related Application to Electromyogram Classification2021

    • Author(s)
      Hayashi Hideaki, Shibanoki Taro, Tsuji Toshio
    • Journal Title

      IEEE Access

      Volume: 9 Pages: 154304-154317

    • DOI

      10.1109/access.2021.3126348

    • Related Report
      2021 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] EMG pattern recognition via Bayesian inference with scale mixture-based stochastic generative models2021

    • Author(s)
      Furui Akira, Igaue Takuya, Tsuji Toshio
    • Journal Title

      Expert Systems with Applications

      Volume: 185 Pages: 115644-115644

    • DOI

      10.1016/j.eswa.2021.115644

    • Related Report
      2021 Annual Research Report
    • Peer Reviewed
  • [Journal Article] A Time-Series Scale Mixture Model of EEG with a Hidden Markov Structure for Epileptic Seizure Detection2021

    • Author(s)
      Furui Akira, Akiyama Tomoyuki, Tsuji Toshio
    • Journal Title

      Proceedings of 43rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society

      Volume: - Pages: 5832-5836

    • DOI

      10.1109/embc46164.2021.9630840

    • Related Report
      2021 Annual Research Report
    • Peer Reviewed
  • [Presentation] Pseudo-Label Learning with Calibrated Confidence Using an Energy-Based Model2024

    • Author(s)
      Masahito Toba, Seiichi Uchida, Hideaki Hayashi
    • Organizer
      IEEE World Congress on Computational Intelligence (WCCI 2024)
    • Related Report
      2023 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Multi-Scale Spatio-Temporal Graph Convolutional Network for Facial Expression Spotting2024

    • Author(s)
      Yicheng Deng, Hideaki Hayashi, Hajime Nagahara
    • Organizer
      International Conference on Automatic Face and Gesture Recognition (FG 2024)
    • Related Report
      2023 Annual Research Report
    • Int'l Joint Research
  • [Presentation] MIDAS: Mixing Ambiguous Data with Soft Labels for Dynamic Facial Expression Recognition2024

    • Author(s)
      Ryosuke Kawamura, Hideaki Hayashi, Noriko Takemura, Hajime Nagahara
    • Organizer
      Winter Conference on Applications of Computer Vision (WACV 2024)
    • Related Report
      2023 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Stochastic Fluctuation in EEG Evaluated via Scale Mixture Model for Decoding Emotional Valence2024

    • Author(s)
      Shunya Fukuda, Akira Furui, Maro Machizawa, and Toshio Tsuji
    • Organizer
      IEEE/SICE International Symposium on System Integration (SII 2024)
    • Related Report
      2023 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Analyzing Font Style Usage and Contextual Factors in Real Images2023

    • Author(s)
      Naoya Yasukochi, Hideaki Hayashi, Daichi Haraguchi, Seiichi Uchida
    • Organizer
      International Conference on Document Analysis and Recognition (ICDAR 2023)
    • Related Report
      2023 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Evaluating Classifier Confidence for Surface EMG Pattern Recognition2023

    • Author(s)
      Akira Furui
    • Organizer
      Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC 2023)
    • Related Report
      2023 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Bayesian Approach for Adaptive EMG Pattern Classification via Semi-supervised Sequential Learning2023

    • Author(s)
      Seitaro Yoneda and Akira Furui
    • Organizer
      IEEE International Conference on Systems, Man, and Cybernetics (SMC 2023)
    • Related Report
      2023 Annual Research Report
    • Int'l Joint Research
  • [Presentation] ランキングと分類による疾病重症度の推定2023

    • Author(s)
      宝満竜一, 内田誠一, 田中聖人, 早志英朗
    • Organizer
      電気・情報関係学会九州支部連合大会
    • Related Report
      2023 Annual Research Report
  • [Presentation] Energy-based modelを用いた信頼度較正と疑似ラベル学習への応用2023

    • Author(s)
      鳥羽真仁, 内田誠一, 早志英朗
    • Organizer
      画像の認識・理解シンポジウム
    • Related Report
      2023 Annual Research Report
  • [Presentation] Mixupを利用した筋電位信号の擬似データ生成と複合動作の識別2023

    • Author(s)
      矢沢 樹, 古居 彬
    • Organizer
      計測自動制御学会システムインテグレーション部門講演会(SI2023)
    • Related Report
      2023 Annual Research Report
  • [Presentation] Deep Bayesian Active Learning to Rank for Endoscopic Image Data2022

    • Author(s)
      Takeaki Kadota, Hideaki Hayashi, Ryoma Bise, Kiyohito Tanaka, and Seiichi Uchida
    • Organizer
      The 26th UK Conference on Medical Image Understanding and Analysis (MIUA2022)
    • Related Report
      2022 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Automated Classification of General Movements in Infants Using a Two-stream Spatiotemporal Fusion Network2022

    • Author(s)
      Yuki Hashimoto, Akira Furui, Koji Shimatani, Maura Casadio, Paolo Moretti, Pietro Morasso, and Toshio Tsuji
    • Organizer
      The 25th International Conference on Medical Image Computing and Computer-assisted Intervention (MICCAI2022)
    • Related Report
      2022 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Combining Generative and Discriminative Models Based on the Gaussian-coupled Softmax Layer2022

    • Author(s)
      Hideaki Hayashi
    • Organizer
      画像の認識・理解シンポジウム (MIRU)
    • Related Report
      2022 Annual Research Report
  • [Presentation] Energy-Based Modelに基づく識別器の信頼度較正2022

    • Author(s)
      鳥羽真仁, 内田誠一, 早志英朗
    • Organizer
      電気・情報関係学会九州支部連合大会
    • Related Report
      2022 Annual Research Report
  • [Presentation] 生体電気信号の尺度混合確率モデルとパターン認識への応用2022

    • Author(s)
      古居 彬
    • Organizer
      2022年電気学会電子・情報・システム部門大会
    • Related Report
      2022 Annual Research Report
  • [Presentation] 非負値行列因子分解を用いた乳児運動解析によるASDリスク評価2022

    • Author(s)
      米井 陸也, 橋本 悠己, 古居 彬, 城明 舜磨, 土居 裕和, 島谷 康司, 辻 敏夫
    • Organizer
      第23回計測自動制御学会システムインテグレーション部門講演会論文集(SI2022)
    • Related Report
      2022 Annual Research Report
  • [Presentation] Meta-learning of Pooling Layers for Few-shot Recognition2021

    • Author(s)
      Takato Otsuzuki, Heon Song, Seiichi Uchida, Hideaki Hayashi
    • Organizer
      画像の認識・理解シンポジウム (MIRU)
    • Related Report
      2021 Annual Research Report
  • [Presentation] 表面筋電位信号のベイズ確率モデルと動作パターン識別2021

    • Author(s)
      古居 彬, 辻 敏夫
    • Organizer
      第60回日本生体医工学会大会プログラム・抄録集
    • Related Report
      2021 Annual Research Report
  • [Presentation] 振幅の確率的変動に着目した非ガウス脳波モデルとてんかん発作自動検出への応用2021

    • Author(s)
      古居 彬, 秋山 倫之, 辻 敏夫
    • Organizer
      第51回日本臨床神経生理学会学術大会 シンポジウム
    • Related Report
      2021 Annual Research Report
  • [Remarks] Hideaki Hayashi

    • URL

      https://sites.google.com/view/hideakihayashi/home

    • Related Report
      2023 Annual Research Report

URL: 

Published: 2021-04-28   Modified: 2025-01-30  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi