RIKEN Center for Advanced Intelligence Project Structured Learning Team
Team Leader: Yoshinobu Kawahara (D.Eng.)
Research Summary
When making predictions based on intelligent information processing such as machine learning, we usually have prior information about structures among variables in data. In our team, we study theories and algorithms for learning with such structural information. Moreover, we conduct applied researches by applying developed algorithms to a variety of scientific and engineering data.
Research Subjects:
- Machine learning with structural prior information
- Development of optimization algorithms for efficient learning
- Learning for analysis of spatiotemporal dynamics
- Application of developed methods to scientific and engineering data
Main Research Fields
- Informatics
Related Research Fields
- Engineering
- Mathematical & Physical Sciences
- Mathematical informatics
- Statistical science
- Intelligent informatics
Keywords
- Machine Learning
- Data Science
- Learning with Prior Information
- Dynamics
- Optimization
Selected Publications
- 1.
Weissenbacher, M., Sinha, S., Garg, A., and Kawahara, Y.:
"Koopman Q-learning: Offline reinforcement learning via symmetries of dynamics"
Proceedings of the 39th International Conference on Machine Learning (ICML'22) (accepted) - 2.
Hashimoto, Y., Ishikawa, I., Ikeda, M., Komura, F., Katsura, T., and Kawahara, Y.:
"Reproducing kernel Hilbert C*-modules and kernel mean embeddings"
Journal of Machine Learning Research, Vol.22, No.267, pp.1-56 (2021) - 3.
Fujii, K., Takeishi, N., Tsutui, K., Fujioka, E., Nishiumi, N., Tanaka, R., Fukushiro, M., Ide, K., Kohno, H., Yoda, K., Takahashi, S., Hiryu, S., and Kawahara, Y.:
"Learning interactions rules from multi-animal trajectories via augmented behavioral models"
Advances in Neural Information Processing Systems 34, pp.11108-11122 (2021) - 4.
Takeishi, N., and Kawahara, Y.:
"Learning dynamics models with stable invariant sets"
Proceedings of the 35th AAAI Conference on Artificial Intelligence, pp.9782-9790 (2021) - 5.
Hashimoto, Y., Ishikawa, I., Ikeda, M., Matsuo, Y. and Kawahara, Y.:
"Krylov subspace method for nonlinear dynamical systems with random noise"
Journal of Machine Learning Research, 21(172): 1-29 (2020) - 6.
Takeishi, N. and Kawahara, Y.:
"Knowledge-based regularization in generative modeling"
Proceedings of the 29th International joint Conference on Artificial Intelligence (IJCAI’20), pp.2390-2396 (2020) - 7.
Takeuchi, N., Yoshida, Y., and Kawahara, Y.:
"Variational inference of penalized regression with submodular functions"
Proceedings of the 35th Conference on Uncertainty in Artificial Intelligence (UAI'19), 443 (2019) - 8.
Fujii, K. and Kawahara, Y.,:
"Dynamic mode decomposition with vector-valued reproducing kernels for extracting dynamical structures among observables"
Neural Networks, 117: 94-103 (2019) - 9.
Ishikawa, I., Fujii, K., Ikeda, M., Hashimoto, Y., and Kawahara, Y.:
"Metric for nonlinear dynamical systems with Koopman operators"
Advances in Neural Information Processing Systems 31, pp.2858-2868 (2018) - 10.
Takeishi, N., Kawahara, Y., and Yairi, T.:
"Learning Koopman invariant subspaces for dynamic mode decomposition"
Advances in Neural Information Processing Systems 30, pp.1130-1140 (2017)
Related Links
Lab Members
Principal investigator
- Yoshinobu Kawahara
- Team Leader
Core members
- Matthias Weissenbacher
- Research Scientist
- Kazu Ghalamkari
- Special Postdoctoral Researcher
- Velmurugan Gandhi
- Postdoctoral Researcher
- Itsushi Sakata
- Postdoctoral Researcher
- Yoshiteru Nishimura
- Technical Staff I
- Keisuke Fujii
- Visiting Scientist
- Shunji Umetani
- Visiting Scientist
- Yuka Hashimoto
- Visiting Scientist
- Takuya Konishi
- Visiting Scientist
- Naoya Takeishi
- Visiting Scientist
Contact Information
744 Motooka, Nishi-ku, Fukuoka-shi, Fukuoka 819-0395 Japan
Email: yoshinobu.kawahara [at] riken.jp