Akira TANIMOTO, Ph.D.

Hello, I’m Akira TANIMOTO. I’m a Research Scientist at NEC Corporation, and I specialize in machine learning.

Research Interest

• Modeling for Decision-Making
  • Causal inference
  • Reinforcement learning
• Sample-efficient ML
  • Side information
  • Transfer learning
• Real-World applications
  • Marketing science: demand forecast and stock optimization, price optimization, and assortment optimization
  • Social infrastructure: railway, satellites, and finance

Academic/professional career summary

[March 2012] B.S. in Department of Engineering, The University of Tokyo

[March 2014] M.S. in Graduate School of Engineering, The University of Tokyo

[April 2014-] Researcher at NEC Corp.

[October 2017-March 2022] Visiting Scientist at RIKEN AIP-NEC Collaboration Center

[September 2021] Ph.D. in Informatics, Kyoto University

Dissertation

1. Akira Tanimoto, Goal-oriented modeling for data-driven decision making, Doctoral dissertation, Kyoto University, 2021.

Publications

Causal inference and application to recommendation:

1. Akira Tanimoto, Tomoya Sakai, Takashi Takenouchi, Hisashi Kashima. Regret Minimization for Causal Inference on Large Treatment Space. International Conference on Artificial Intelligence and Statistics (AISTATS), 2021.
2. Akira Tanimoto, Tomoya Sakai, Takashi Takenouchi, Hisashi Kashima. Causal Combinatorial Factorization Machines for Set-wise Recommendation. In Proceedings of the 25th Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD), 2021.

Privileged (side) information:

3. Akira Tanimoto, So Yamada, Takashi Takenouchi, Masashi Sugiyama, Hisashi Kashima. Improving imbalanced classification using near-miss instances, Expert Systems with Applications, 2022.
4. Shogo Hayashi, Akira Tanimoto, Hisashi Kashima, Long-Term Prediction of Small Time-Series Data Using Generalized Distillation, In Proceedings of the 2019 International Joint Conference on Neural Networks (IJCNN), 2019.

Reinforcement learning application:

2. Akira Tanimoto, Combinatorial Q-Learning for Condition-Based Infrastructure Maintenance, in IEEE Access, vol. 9, pp. 46788-46799, 2021.

Please refer Japanese page for publications in Japanese.

Please also refer to Google scholar for more info.

Contact

a.tanimoto[at]nec.com

or my personal email