Toru Fujino (藤野 暢)

I'm a machine learning engineer at Idein Inc., Japan.

Email: toru.fb34 (at)

Research interests

  • Machine learning / deep learning for nontransitive games
  • Deep learning for medical image analysis
  • Deep learning for natural language processing / generation
  • Career


  • Machine learning engineer, Idein Inc. (Apr. 2021 - Present)
  • Researcher / project researcher, National Cancer Center Hospital East (Jan. 2020 - Mar. 2021)
  • Data scientist, IGPI Business Analytics & Intelligence, Inc. (Dec. 2015 - Apr. 2018, Oct. 2018 - Nov. 2019)
  • Education

  • Ph.D. student (Withdrawn), Graduate School of Frontier Sciences, The University of Tokyo, Japan (Apr. 2016 - Mar. 2021)
  • Masters in Environmental Studies, Graduate School of Frontier Sciences, The University of Tokyo, Japan (Mar. 2016)
  • Bachelor in Economics, Department of Economics, Yokohama National University, Japan (Mar. 2014)
  • Research


    1. Daichi Kitaguchi, Toru Fujino, Nobuyoshi Takeshita, Hiro Hasegawa, Kensaku Mori, and Masaaki Ito. Limited generalizability of single deep neural network for surgical instrument segmentation in different surgical environments. Scientific Reports. 2022. [Paper]
    2. Toru Fujino and Yu Chen. Effects of Network Structure on the Performance of a Modeled Traffic Network under Drivers' Bounded Rationality. Physica A: Statistical Mechanics and its Applications. 2020. [Paper] [Code]

    Conference proceeding

    1. Masaru Isonuma, Toru Fujino, Ichiro Sakata, Yutaka Matsuo, and Junichiro Mori. Extractive Summarization Using Multi-Task Learning with Document Classification. Conference on Empirical Methods in Natural Language Processing (EMNLP). 2017. [Paper]

    Conference presentations (non-archival)

    1. Toru Fujino and Yu Chen. Do Stylized Facts in Financial Markets Collapse Due to the Artificial Intelligence Analysis of Other Traders Bahavior? Conference on Complex Systems (CCS). 2019.
    2. Toru Fujino and Yu Chen. Effects of Network Structures on the Performance of a Modeled Traffic Network with Drivers’ Preference Heterogeneity. The 3rd Annual International Conference on Computational Social Science (IC2S2). 2017.
    3. Toru Fujino, Kangwei Chen, and Yu Chen. Study on Transport Costs across Network Using a Minority Game Model with Imperfect Information. Social Modeling and Simulations + Econophysics Colloquim (SMSEC). 2014.



  • Scholarship for advanced graduate student in artificial intelligence by Toyota & Dowango (トヨタ・ドワンゴ高度人工知能人材奨学金) (Apr. 2017 - Mar. 2018)
  • Teaching

  • Deep learning basics at the University of Tokyo (2016 - 2019, Worked as a teaching staff) [link]
  • Deep Learning for natural language processing summer school at the University of Tokyo (Aug. 2018 - Sep. 2018, Worked as a teaching staff) [link]
  • Blog/Articles (in Japanese)

  • ゲーム理論でLLMの信頼性アップ 生成AIの課題、解決へ前進. 東大松尾ゼミの「深層学習」研究会 第59回. 日経クロストレンド. 2024. [link]
  • ゲームにおけるNon-transitivityについて. Idein株式会社note. 2022. [link]
  • 1層でも高精度 不動点を用いたディープラーニング「DEQ」とは. 東大松尾ゼミの「深層学習」研究会 第25回. 日経クロストレンド. 2021. [link]
  • 協調AIなら解ける? 経済学のジレンマ「共有地の悲劇」. 東大松尾ゼミの「深層学習」研究会 第13回. 日経クロストレンド. 2020. [link]
  • Skills

  • Softwares: Python (intermediate), C/C++ (lower-intermediate), Julia (beginner) Rust (absolute beginner)
  • Languages: Japanese (native), English (upper-intermediate. TOEFL iBT: 96, Sep. 2016)
  • Links

  • Google Scholar: Toru Fujino
  • SlideShare: torufujino
  • GitHub: toru34
  • GitLab: toru34
  • LinkedIn: toru34
  • Hobby

  • CrossFit: Toru Fujino