Publications and Preprints 🔗

  1. Ishikawa, I.*, Teshima, T.*, Tojo, K., Oono, K., Ikeda, M., Sugiyama, M.,
    Universal approximation property of invertible neural networks. arXiv:2204.07415 [cs.LG], 2022.
    * Equal contribution.

    Preprint (arXiv)
  2. Lu, N., Zhang, T., Fang, T., Teshima, T., Sugiyama, M.,
    Rethinking importance weighting for transfer learning. arXiv:2112.10157 [cs.LG], 2021.
    Preprint (arXiv)
  3. Oishi, A., Teshima, T., Kojima, N., Kiriha, M., Kojima, N., Sasaki, T., Takahira, K., Takeuchi, T., Tajima, K., Noda, C., Hirose, H., and Yamanaka, S.,
    Forecasting internally displaced people's movements with artificial intelligence (Book Chapter). Digital Innovations, Business and Society in Africa: New Frontiers and a Shared Strategic Vision (Springer), 2022.
    Book series: Advances in Theory and Practice of Emerging Markets". Book editors: Richard Boateng, Sheena Lovia Boateng, Thomas Anning-Dorson and Longe Olumide Babatope"

    Chapter page (Springer) Book (Springer)
  4. Teshima, T., Tojo, K., Ikeda, M., Ishikawa, I., and Oono, K.,
    Universal approximation property of neural ordinary differential equations. NeurIPS2020 Workshop: Differential Geometry meets Deep Learning, 2020 (DiffGeo4DL).
    preprint (arXiv) slides venue and video
  5. Teshima, T.*, Ishikawa, I.*, Tojo, K., Oono, K., Ikeda, M., and Sugiyama, M.,
    Coupling-based invertible neural networks are universal diffeomorphism approximators. Advances in Neural Information Processing Systems 33, 2020 (NeurIPS 2020).
    * Equal contribution. Oral presentation (one of the 105 orals among the 1900 accepted papers; paper acceptance rate 20.1%, oral acceptance rate 1.1%).

    Proceedings code (figure) slides Conference session
  6. Fujisawa, M., Teshima, T., and Sato, I.,
    γ-ABC: Outlier-robust approximate Bayesian computation based on a robust divergence estimator. The 24th International Conference on Artificial Intelligence and Statistics, 2021 (AISTATS 2021).
  7. Kato, M. and Teshima, T.,
    Non-negative Bregman divergence minimization for deep direct density ratio estimation. Proceedings of the 38th International Conference on Machine Learning, 2021 (ICML 2021).
    Proceedings Press release
  8. Kato, M., Teshima, T., and Honda, J.,
    Learning from positive and unlabeled data with a selection bias. Seventh International Conference on Learning Representations, 2019 (ICLR 2019).
  9. Teshima, T., Xu, M., Sato, I., and Sugiyama, M.,
    Clipped matrix completion: a remedy for ceiling effects. Thirty-Third AAAI Conference on Artificial Intelligence, 2019 (AAAI-19).
    paper supplementary code slides