Interests

• Machine learning (probabilistic models, Gaussian processes, deep learning, etc.)

• Physics-informed machine learning

• Data mining (spatio-temporal data analysis, user modeling, etc.)

Publications

Conference papers

1. Yusuke Tanaka, Learning Hamiltonian Dynamics Under Uncertainty via Symplectic Gaussian Processes, SCML, 2024 (to appear).

2. Yusuke Tanaka, Tomoharu Iwata, Naonori Ueda, Symplectic Spectrum Gaussian Processes: Learning Hamiltonians from Noisy and Sparse Data, NeurIPS, 2022.

3. Maya Okawa, Tomoharu Iwata, Yusuke Tanaka, Hiroyuki Toda, Takeshi Kurashima, and Hisashi Kashima, Dynamic Hawkes Processes for Discovering Time-evolving Communities' States behind Diffusion Processes, KDD, Virtual Conference, pp. 1276 - 1286, 2021. (acceptance rate: 238/1541 = 15.4%)

4. Yasunori Akagi, Takuya Nishimura, Yusuke Tanaka, Takeshi Kurashima, Hiroyuki Toda, Exact and Efficient Inference for Collective Flow Diffusion Model via Minimum Convex Cost Flow Algorithm, AAAI, New York, United States, pp. 3163 - 3170, 2020. (acceptance rate: 1591/7737 = 20.6%) 

5. Yoshiaki Takimoto, Yusuke Tanaka, Takeshi Kurashima, Shuhei Yamamoto, Maya Okawa, Hiroyuki Toda, Predicting Traffic Accidents with Event Recorder Data, ACM SIGSPATIAL Workshop on Prediction of Human Mobility (PredictGIS), Chicago, United States, pp. 11 - 14, 2019. 

6. Yusuke Tanaka, Toshiyuki Tanaka, Tomoharu Iwata, Takeshi Kurashima, Maya Okawa, Yasunori Akagi, Hiroyuki Toda, Spatially Aggregated Gaussian Processes with Multivariate Areal Outputs, NeurIPS, Vancouver, Canada, pp. 3000 - 3010, 2019. (acceptance rate: 1428/6743 = 21.2%) [poster], [code]

7. Maya Okawa, Tomoharu Iwata, Takeshi Kurashima, Yusuke Tanaka, Hiroyuki Toda, Naonori Ueda, Deep Mixture Point Processes: Spatio-temporal Event Prediction with Rich Contextual Information, KDD, Anchorage, United States, pp. 373 - 383, 2019. (acceptance rate: 170/1200 = 14.2%)

8. Yusuke Tanaka, Tomoharu Iwata, Toshiyuki Tanaka, Takeshi Kurashima, Maya Okawa, Hiroyuki Toda, Refining Coarse-grained Spatial Data using Auxiliary Spatial Data Sets with Various Granularities, AAAI, Honolulu, United States, pp. 5091 - 5100, 2019. (acceptance rate: 1150/7095 = 16.2%) [slide, poster]

9. Yusuke Tanaka, Tomoharu Iwata, Takeshi Kurashima, Hiroyuki Toda, Naonori Ueda, Estimating Latent People Flow without Tracking Individuals, IJCAI, Stockholm, Sweden, pp. 3556 - 3563, 2018. (acceptance rate: 710/3470 = 20.5%) [slide, poster]

10. Yusuke Tanaka, Takeshi Kurashima, Yasuhiro Fujiwara, Tomoharu Iwata, Hiroshi Sawada, Inferring Latent Triggers of Purchases with Consideration of Social Effects and Media Advertisements, WSDM, San Francisco, United States, pp. 543 - 552, 2016. (acceptance rate: 67/368 = 18.2%) [slide]

11. Naonori Ueda, Yusuke Tanaka, Akinori Fujino, Robust Naive Bayes Combination of Multiple Classifications, FMI, Fukuoka, Japan, pp. 141 - 155, 2013.

12. Takuya Kitamura, Shigeo Abe, Yusuke Tanaka, Multiple Nonlinear Subspace Methods Using Subspace-based Support Vector Machines, ICMLA, Honolulu, United States, pp. 358 - 363, 2011.

Journal papers

1. Maya Okawa, Tomoharu Iwata, Yusuke Tanaka, Takeshi Kurashima, Hiroyuki Toda, Hisashi Kashima, Context-aware spatio-temporal event prediction via convolutional Hawkes processes, Machine Learning, Mar. 2022. 

2. Tomoharu Iwata, Yusuke Tanaka, Few-shot learning for spatial regression via neural embedding-based Gaussian processes, Machine Learning, Nov. 2021. 

3. Maya Okawa, Tomoharu Iwata, Takeshi Kurashima, Yusuke Tanaka, Hiroyuki Toda, Naonori Ueda, and Hisashi Kashima, Deep Mixture Point Processes: Spatio-temporal Event Prediction with External Factor, Journal of Japanese Society for Artificial Intelligence, Vol.36, No.5, p.C-L37_1-10, Sep. 2021 (in Japanese). 

4. Yusuke Tanaka, Tomoharu Iwata, Takeshi Kurashima, Hiroyuki Toda, Naonori Ueda, Toshiyuki Tanaka, Time-delayed Collective Flow Diffusion Models for Inferring Latent People Flow from Aggregated Data at Limited Locations, Artificial Intelligence, Vol. 292, 103430, Mar. 2021.

5. Yoshiaki Takimoto, Yusuke Tanaka, Takeshi Kurashima, Shuhei Yamamoto, Maya Okawa, Hiroyuki Toda, Predicting Traffic Accidents with Dashboard Cameras, IPSJ-TOD, Vol. 14, No. 1, pp. 1 - 7, Jan. 2021. (in Japanese)

6. Maya Okawa, Yusuke Tanaka, Takeshi Kurashima, Hiroyuki Toda, Tomohiro Yamada, Marked Temporal Point Processes for Trip Demand Prediction in Bike Sharing Systems, IEICE Trans. Inf. & Syst., Vol. E102-D, No. 9, pp. 1635-1643, Sep. 2019.

7. Yusuke Kawai, Yusuke Tanaka, Hiroyuki Toda, Yoshiharu Ishikawa, Estimating People Flow from a Large Amount of Aggregated Data with a Few Tracking Data, DBSJ Japanese Journal, Vol.17-J, No. 7, Mar. 2019. (in Japanese)

8. Yusuke Tanaka, Takeshi Kurashima, Yasuhiro Fujiwara, Tomoharu Iwata, Hiroshi Sawada, Identifying Latent Triggers of Purchases with Consideration of Social Effects and Media Advertisements, IPSJ Journal, Vol. 58, No. 2, pp. 580 - 593, Feb. 2017. (in Japanese)

9. Yusuke Tanaka, Naonori Ueda, Toshiyuki Tanaka, Bayesian Classifier Based on Class-specific Feature Selection, IEICE Trans. Inf. & Syst., Vol. J96-D, No. 11, pp. 2755 − 2764, Nov. 2013. (in Japanese)

Preprint

• Yusuke Tanaka, Takaharu Yaguchi, Tomoharu Iwata, Naonori Ueda, Neural Operators Meet Energy-based Theory: Operator Learning for Hamiltonian and Dissipative PDEs, arXiv:2402.09018, 2024.

• Yoshiaki Takimoto, Yusuke Tanaka, Tomoharu Iwata, Maya Okawa, Hideaki Kim, Hiroyuki Toda, Takeshi Kurashima, Meta-learning for Neural Network-based Temporal Point Processes, arXiv:2401.15846, 2024.

• Takeshi Koshizuka, Masahiro Fujisawa, Yusuke Tanaka, Issei Sato, Initialization Bias of Fourier Neural Operator: Revisiting the Edge of Chaos, arXiv:2310.06379, 2023.

• Tomoharu Iwata, Yusuke Tanaka, Naonori Ueda, Meta-learning of Physics-informed Neural Networks for Efficiently Solving Newly Given PDEs, arXiv:2310.13270, 2023.

• Yusuke Tanaka, Toshiyuki Tanaka, Tomoharu Iwata, Takeshi Kurashima, Maya Okawa, Yasunori Akagi, Hiroyuki Toda, Aggregated Multi-output Gaussian Processes with Knowledge Transfer Across Domains, arXiv:2206.12141, 2022.  [code]

• Yasunori Akagi, Yusuke Tanaka, Tomoharu Iwata, Takeshi Kurashima, Hiroyuki Toda, Probabilistic Optimal Transport based on Collective Graphical Models, arXiv:2006.08866, 2020.

Educational activity

• Advisor, NTT Summer Internship, 2017.

• Visiting Associate Professor, Graduate School of Science and Technology, Nara Institute of Science and Technology (NAIST), Apr. 2021 -

Research activities

Invited talk

• Yusuke Tanaka, Inferring Latent People Flows from Aggregated Data in Real-world Settings, ISCIE SSS, 2018.

Professional activities

• Reviewer, Transactions on Machine Learning Research (TMLR), 2022-

• Program Committee Member, WSDM, 2022, 2023.

• Reviewer, ICLR, 2021, 2022, 2023.

• Reviewer, ICML, 2020 (top reviewers), 2021, 2022, 2023, 2024.

• Program Committee Member, IJCAI, 2020, 2022.

• Program Committee Member, AAAI, 2024.

• Reviewer, NeurIPS, 2019 (top reviewers), 2020, 2021, 2022, 2023.

Biography

• Dec. 2020 - Present, Research scientist at NTT Communication Science Laboratories, NTT Corporation, Japan

• Apr. 2013 - Nov. 2020, Researcher at NTT Service Evolution Laboratories, NTT Corporation, Japan

• Apr. 2017 - Mar. 2020, Doctor of Informatics, Kyoto University (Supervisor: Dr. Toshiyuki Tanaka)

• Apr. 2011 - Mar. 2013, Master of Informatics, Kyoto University (Supervisor: Dr. Naonori Ueda, Dr. Toshiyuki Tanaka)

• Apr. 2007 – Mar. 2011, Bachelor of Engineering, Kobe University (Supervisor: Dr. Shigeo Abe)

Thesis

• Yusuke Tanaka, Probabilistic Models for Spatially Aggregated Data, Doctoral Thesis, Graduate School of Informatics, Kyoto University, 2020. [slide]

• Yusuke Tanaka, Bayesian Classifier based on Class-Specific Feature Selection, Master's Thesis, Graduate School of Informatics, Kyoto University, 2013.

Society membership

• The Institute of Electronics, Information and Communication Engineers (IEICE)

• The Japanese Society for Artificial Intelligence (JSAI)