Assistant Professor

Mathematical Informatics 4th laboratory, Department of Mathematical Informatics, The University of Tokyo

Email: matsuda[at]mist.i.u-tokyo.ac.jp

- 2012.3: Bachelor of Engineering from Faculty of Engineering, University of Tokyo
- 2014.3: Master of Information Science and Technology from Graduate School of Information Science and Technology, University of Tokyo
- 2017.3: Ph.D of Information Science and Technology from Graduate School of Information Science and Technology, University of Tokyo
- 2014.4-2017.3: JSPS Research Fellow (DC1)
- 2017.4-: Assistant Professor, University of Tokyo
- 2017.6-2018.3: Visiting Researcher, RIKEN Brain Science Institute
- 2018.4-: Visiting Researcher, RIKEN Center for Brain Science

- T. Matsuda and T. Matsuo. A new geometric integration approach based on local invariants.
*JSIAM Letters*, Vol. 5, pp. 37--40, 2013. (link) - T. Matsuda and F. Komaki. Singular value shrinkage priors for Bayesian prediction.
*Biometrika*, Vol. 102, pp. 843--854, 2015. (link) - T. Matsuda and W. E. Strawderman. Pitman closeness properties of point estimators and predictive densities with parametric constraints.
*Statistics & Probability Letters*, Vol. 116, pp. 101--106, 2016. (link) - T. Matsuda and W. E. Strawderman. Pitman closeness properties of Bayes shrinkage procedures in estimation and prediction.
*Statistics & Probability Letters*, Vol. 119, pp. 21--29, 2016. (link) - T. Matsuda and F. Komaki. Time series decomposition into oscillation components and phase estimation.
*Neural Computation*, Vol. 29, pp. 332--367, 2017. (link, MATLAB code) - T. Matsuda, K. Kitajo, Y. Yamaguchi and F. Komaki. A point process modeling approach for investigating the effect of online brain activity on perceptual switching.
*NeuroImage*, Vol. 152, pp. 50--59, 2017. (link) - T. Matsuda and F. Komaki. Multivariate time series decomposition into oscillation components.
*Neural Computation*, Vol. 29, pp. 2055--2075, 2017. (link, MATLAB code) - M. Tsuji, T. Kawasaki, T. Matsuda, S. Gojo and J. Takeuchi. Sexual dimorphisms of mRNA and miRNA in human/murine heart disease.
*PLOS ONE*, Vol. 12, e0177988, 2017. (link) - T. Koyama, T. Matsuda and F. Komaki. Minimax estimation of quantum states based on the latent information priors.
*Entropy*, Vol. 19, 618, 2017. (link) - T. Matsuda and W. E. Strawderman. Improved loss estimation for a normal mean matrix.
*Journal of Multivariate Analysis*, Vol. 169, pp. 300--311, 2019. (link, METR) - T. Matsuda and A. Hyvarinen. Estimation of Non-Normalized Mixture Models. 22nd International Conference on Artificial Intelligence and Statistics (AISTATS 2019). (link)
- T. Matsuda and F. Komaki. Empirical Bayes matrix completion.
*Computational Statistics & Data Analysis*, accepted. (arXiv) - Y. Maruyama, T. Matsuda and T. Onishi. Harmonic Bayesian prediction under alpha-divergence.
*IEEE Transactions on Information Theory*, accepted. - T. Matsuda. Statistical analysis of
*kimariji*in competitive karuta (in Japanese). submitted. - T. Matsuda and F. Komaki. Improving on singular value shrinkage priors and block-wise Stein priors. (METR)
- T. Ohki
^{*}, T. Matsuda^{*}, A. Gunji, Y. Takei, R. Sakuma, H. Takahashi, Y. Kaneko, M. Inagaki, T. Hanakawa and K. Hiraki. Timing of phase amplitude coupling in the temporal pole is essential for neuronal and functional maturation in adolescence. submitted. (*: co-first) - T. Matsuda and A. Takemura. Game-theoretic derivation of upper hedging prices of multivariate contingent claims and submodularity. submitted. (arXiv)
- M. Uehara, T. Matsuda and F. Komaki. Analyisis of noise contrastive estimation from the perspective of asymptotic variance. submitted. (arXiv)
- M. Uehara, T. Kanamori, T. Takenouchi and T. Matsuda. A unified estimation framework for unnormalized models with statistical efficiency. submitted. (arXiv)
- M. Uehara, T. Matsuda and J. K. Kim. Imputation estimators for unnormalized models with missing data. submitted. (arXiv)
- T. Matsuda and W. E. Strawderman. Predictive density estimation under the Wasserstein loss. submitted. (arXiv)

- Time series decomposition into oscillation components and phase estimation (univariate time series, multivariate time series)
- Empirical Bayes matrix completion

- 2014.7
__T. Matsuda__and F. Komaki. Singular value shrinkage priors for Bayesian prediction. ims-APRM 2014, Taipei. - 2015.6
__T. Matsuda__and F. Komaki. Singular value shrinkage priors for Bayesian prediction. 11th International Workshop on Objective Bayes Methodology (O-Bayes15), Valencia. - 2015.8
__T. Matsuda__, K. Kitajo, Y. Yamaguchi and F. Komaki. Point process modeling of perceptual switching. SAMSI Program on Challenges in Computational Neuroscience (CCNS) Opening Workshop, Duke. - 2016.6
__T. Matsuda__and F. Komaki. Decomposing time series into oscillation components with random frequency modulation. ims-APRM 2016, Hong Kong. - 2016.12
__T. Matsuda__and F. Komaki. Improvement of singular value shrinkage priors and block-wise Stein priors. 9th International Conference of the ERCIM WG on Computational and Methodological Statistics (CMStatistics 2016), Seville. - 2018.6 T. Matsuda. Minimax adaptive reduced-rank regression. 5th IMS Asia Pacific Rim Meeting (ims-APRM 2018), Singapore.
- 2018.7
__T. Ohki__, T. Matsuda, A. Gunji, Y. Takei, R. Sakuma, H. Takahashi, Y. Kaneko, M. Inagaki, T. Hanakawa and K. Hiraki. Timing of phase amplitude coupling in the temporal pole is essential for neuronal and functional maturation in adolescence. 11th FENS Forum of Neuroscience, Berlin. - 2018.7
__T. Matsuda__, F. Homae, H. Watanabe, G. Taga and F. Komaki. Statistical verification of the common oscillatory behaviors in oxy-Hb and deoxy-Hb time series. Toward Understanding ``INDIVIDUALITY", Kyoto. - 2018.10
__T. Matsuda__, F. Homae, H. Watanabe, G. Taga and F. Komaki. Statistical verification of the common oscillatory behaviors in oxy-Hb and deoxy-Hb time series. fNIRS 2018, Tokyo. - 2019.2 T. Matsuda and
__Y. Miyatake__. Reducing the effect of discretization errors in estimating ODE models by iteratively reweighted least squares. ANZIAM 2019, Nelson. - 2019.3 T. Matsuda and
__Y. Miyatake__. Estimating ODE models by iteratively reweighted least squares. A3 Workshop on fluid dynamics and related topics, Kobe. - 2019.4 T. Matsuda. Singular value shrinkage priors and empirical Bayes matrix completion. New and Evolving Roles of Shrinkage in Large-Scale Prediction and Inference, Banff. (video)
- 2019.4
__T. Matsuda__and A. Hyvarinen. Estimation of Non-Normalized Mixture Models. 22nd International Conference on Artificial Intelligence and Statistics (AISTATS 2019), Okinawa. - 2019.6 T. Matsuda and
__Y. Miyatake__. Quantifying and reducing forward uncertainty in estimating parameters of initial value problems. EASIAM 2019, Wuhan. - 2019.6 T. Matsuda. TBA. Symposium in Memory of Charles Stein, Singapore.
- 2019.7 T. Matsuda. Singular value shrinkage priors for Bayesian prediction. EAC-ISBA 2019, Kobe.
- 2019.7
__T. Matsuda__and Y. Miyatake. TBA. SciCADE 2019, Austria.

- Dean's award from Faculty of Engineering, University of Tokyo (2012.3)
- O-Bayes15 Jeffreys Poster Excellence Award (2015.6)
- Best Presentation Award, Japanese Joint Statistical Meeting (2015.9)
- Dean's award from Graduate School of Information Science and Technology, University of Tokyo (2017.3)

- Mathematical Informatics 4th laboratory
- RIKEN CBS The University of Tokyo Collaboration Unit
- RIKEN Brain Science Training Program (I attended this course in 2013-14)