教員一覧 List of faculty members
島内 宏和 准教授
SHIMAUCHI, Hirokazu Associate Professor
島内 宏和 准教授
SHIMAUCHI Hirokazu所属：
複雑系知能学科、 複雑系情報科学領域（大学院）、 知能情報科学領域（大学院）
研究分野
機械学習前職・前歴
山梨英和大学、東京工業大学（現 : 東京科学大学）、公益財団法人 東京財団政策研究所、八戸工業大学担当科目（学部）
機械学習Ⅰ、機械学習Ⅱ、データサイエンス基礎、データサイエンス応用、オペレーティングシステム、数学総合演習Ⅱ、システム情報科学実習、モデリング入門1担当科目（大学院）
数理解析特論学位
博士（情報科学）、東北大学
SHIMAUCHI, Hirokazu Associate Professor
Affiliation：
Department of Complex and Intelligent Systems、 Complex System Information Science Field
Research Fields
Machine LearningAcademic Background
Yamanashi Eiwa College, Tokyo Institute of Technology, Tokyo Foundation for Policy Research, Hachinohe Institute of TechnologySubjects in Charge (Undergraduate)
Machine Learning 1, Machine Learning 2, Basics of Data Science, Applied Data Science, Operating Systems, Mathematics Practice 2, Systems Information Science Practice, Introduction to ModelingSubjects in Charge (Graduate School)
Advanced Topics in Mathematical AnalysisDegree
Ph.D. in Information Sciences, Tohoku University関連リンク
研究内容
データに潜む規則性を見出す機械学習の手法の構築と応用に取り組んでいます。近年は、擬等角写像論に基づく表現学習・外れ値検出、深層学習における新しい活性化・表現変換手法に加え、量子コンピュータを活用した非可換量子ダイナミクスに基づく特徴表現学習にも取り組んでいます。また、社会科学や流体解析などの分野に対する機械学習の応用研究についても、他分野の研究者と連携しながら進めています。
研究の魅力
「機械」がデータから「学習」する仕組みに関するアルゴリズムを、新しい発想とアプローチで自由に考えていくことができる点が魅力だと思っています。既存の枠組みに捉われることなく、「0を1にする」ような研究を目指しています。
実績
- Best Paper Candidates (上位3編), IEEE/ACM International Conference on Big Data Computing, Applications and Technologies (BDCAT 2019), 2019年 2019 .
- Best Paper Candidates, 15th International Joint Conference on Computational Intelligence (IJCCI 2023): Neural Computation Theory and Applications (NCTA 2023), 2023年 2023 .
- 青森県工業技術教育振興会奨励賞受賞, 2024年 2024 .
主な著作・論文
- H. Shimauchi, Noncommutative Quantum Dynamics for Feature Representation Learning, Accepted for publication in IEEE Access, 2026 (査読あり).
- H. Shimauchi, Geometry-Aware Stochastic Structured Channel Mixing for Deep Neural Networks, Accepted for publication in the Proceedings of 6th International Conference on Image Processing and Vision Engineering, Communications in Computer and Information Science, Springer, 2026 (査読あり).
- 島内宏和, 擬等角写像の数値的構成法と機械学習への応用, 電子情報通信学会 基礎・境界ソサイエティ Fundamentals Review, 19(2), 97-104, 2025 (招待あり).
- H. Shimauchi, Quasiconformal Extension-Based Unsupervised Representation Learning and Application to Semi-supervised Outlier detection, Computational Intelligence, 187–210, 2025 (査読あり).
- H. Shimauchi, Unsupervised Representation Learning by Quasiconformal Extension, In Proceedings of the 15th International Joint Conference on Computational Intelligence, 1, 440-449, 2023 (査読あり).
- H. Shimauchi, An Activation Function with Probabilistic Beltrami Coefficient for Deep Learning, In Proceedings of the 14th International Conference on Agents and Artificial Intelligence, 3, 613-620, 2022 (査読あり).
- H. Shimauchi, Improving Supervised Outlier Detection by Unsupervised Representation Learning and Generative Adversarial Networks, In Proceedings of the 4th International Conference on Information Science and Systems, 22-27, 2021 (査読あり).
- S. Kato, T. Nakanishi, B. Ahsan, H. Shimauchi, Time-series topic analysis using singular spectrum transformation for detecting political business cycles, Journal of Cloud Computing, 10, 21, 1-16, 2021 (査読あり).
- S. Kato, T. Nakanishi, H. Shimauchi, B. Ahsan, Topic Variation Detection Method for Detecting Political Business Cycles, In Proceedings of the 6th IEEE/ACM International Conference on Big Data Computing, Applications and Technologies, 85-93, 2019 (査読あり).
- 東北大学学際科学フロンティア研究所「百科繚覧」編集委員会, 百科繚覧 Vol 1 : 若手研究者が挑む学際フロンティア（担当箇所：第3章）, 東北大学出版会, 2019.
（研究室の卒業研究生・大学院生との共同研究） - Y. Kanda, H. Shimauchi, Y. Katori, Knowledge Distillation From TSMixer to Echo StateNetwork for Multivariate Time-Series Forecasting, IEICE GlobalNet Workshop 2026, 2026.
- 今 壮平, 島内 宏和, エッジ環境に適した教師なし異常検知のための深さ方向分離畳み込みと知識蒸留による軽量深層学習モデル, 電子情報通信学会総合大会 (センサネットワークとモバイルインテリジェンス), 2026.
- 多田 瑛貴, 島内 宏和, 野村 怜佳, 森口 周二, 寺田 賢二郎, 高瀬 慎介, 外里 健太, 沖合津波観測データの潜在表現学習とPhysics-Informed Neural Networkによる津波の即時予測, 電子情報通信学会総合大会 (人工知能と知識処理), 2026.
Research Contents
I am conducting research on constructing and applying machine learning methods to uncover hidden patterns in data. Specifically, I am working on representation learning techniques that derive useful features for prediction from data. I am also developing methods to detect outliers that significantly differ from overall data trends. Additionally, I collaborate with researchers from various disciplines on applying machine learning to issues in the social sciences and other fields.
Attractive Factors of My Research
I am captivated by the point to freely constructing algorithms related to the mechanisms by which machines learn automatically, using new ideas and approaches. I aim to conduct research that transforms zero into one, without being confined by existing frameworks.
Achievement
- Best Paper Candidates, IEEE/ACM International Conference on Big Data Computing, Applications and Technologies (BDCAT 2019), 2019.
- Best Paper Candidates, 15th International Joint Conference on Computational Intelligence (IJCCI 2023): Neural Computation Theory and Applications (NCTA 2023), 2023.
Major Books and Papers
- H. Shimauchi, Unsupervised Representation Learning by Quasiconformal Extension, In Proceedings of the 15th International Joint Conference on Computational Intelligence, 1, 440-449, 2023.
- H. Shimauchi, An Activation Function with Probabilistic Beltrami Coefficient for Deep Learning, In Proceedings of the 14th International Conference on Agents and Artificial Intelligence, 3, 613-620, 2022.
- H. Shimauchi, Improving Supervised Outlier Detection by Unsupervised Representation Learning and Generative Adversarial Networks, In Proceedings of the 4th International Conference on Information Science and Systems, 22-27, 2021.
- S. Kato, T. Nakanishi, B. Ahsan, H. Shimauchi, Time-series topic analysis using singular spectrum transformation for detecting political business cycles, Journal of Cloud Computing, 10, 21, 1-16, 2021.
- S. Kato, T. Nakanishi, H. Shimauchi, B. Ahsan, Topic Variation Detection Method for Detecting Political Business Cycles, In Proceedings of the 6th IEEE/ACM International Conference on Big Data Computing, Applications and Technologies, 85-93, 2019.
