Takaharu YAGUCHI 
Takaharu YAGUCHI
Department of Mathematics,
Graduate School of Science, Kobe University 
Division : Applied Mathematics
Professor 
Building B, Room 322

Research Field :
Development and theoretical analysis of information technologies, particularly physical simulation and deep learning 

Research Summary :
I am addressing various real world problems through a mathematical approach.
Currently, I am focusing on the development of methods of deep geometric scientific computation, in which deep learning,
geometric mechanics and physical simulation techniques are combined.


Primary Publications : 

T. Matsubara and T. Yaguchi, FINDE: Neural Differential Equations for Finding and Preserving Invariant Quantities,
Proc. of The Eleventh International Conference on Learning Representations (ICLR2023), Kigali, May 2023.

Y. Chen, T. Matsubara, and T. Yaguchi, KAM Theory Meets Statistical Learning Theory: Hamiltonian Neural Networks with NonZero Training Loss,
Proc. of The ThirtySixth AAAI Conference on Artificial Intelligence (AAAI2022, oral), Virtual, Feb. 2022.

Y. Chen, T. Matsubara, and T. Yaguchi, Neural Symplectic Form: Learning Hamiltonian Equations on General Coordinate Systems, Advances in Neural Information Processing Systems 34 (NeurIPS2021, spotlight), Virtual, Dec. 2021.

T. Matsubara, A. Ishikawa, T. Yaguchi, Deep Energybased Modeling of DiscreteTime Physics, Advances in Neural Information Processing Systems (NeurIPS, oral), Virtual, Dec. 2020.

M. Komatsu, T. Yaguchi, K. Nakajima, Algebraic approach towards the exploitation of "softness": the inputoutput equation for morphological computation, International Journal of Robotics Research, 40 (2021) 99118

K. Masumoto, T. Yaguchi, H. Matsuda, H. Tani, K. Totsuka, N. Kondo and S. Okada, Measurement and visualization of facetoface interaction among communitydwelling elderly persons using wearable sensors, Geriatrics and Gerontology International, 17 (2017) 17521758.

T. Yaguchi, A Lagrangian Approach to Deriving EnergyPreserving Numerical Schemes for the EulerLagrange Partial Differential Equations, M2AN, 47 (2013) 14931513.

T. Yaguchi, T. Matsuo and M. Sugihara, "The Discrete Variational Derivative Method Based on Discrete Differential Forms," J. Comput. Phys., 231 (2012) 39633986.


