Research

Working Papers

  1. Masahiro Kato, Kyohei Okumura, Takuya Ishihara, and Toru Kitagawa
    Adaptive Experimental Design for Policy Learning [arXiv]
  2. Masahiro Kato, Kei Nakagawa, Kenshi Abe, Tetsuro Morimura, and Kentaro Baba
    Direct Expected Quadratic Utility Maximization for Mean-Variance Controlled Reinforcement Learning
    First draft: 29 Sept 2020; updated 3 Apr 2021. [arXiv]
  3. Masahiro Kato, Kota Matsui, and Ryo Inokuchi
    Double Debiased Covariate Shift Adaptation Robust to Density-Ratio Estimation [arXiv]
    Submitted.
  4. Masahiro Kato
    Adaptive Generalized Neyman Allocation: Local Asymptotic Minimax Optimal Best Arm Identification [arXiv]
    Submitted.
  5. Kaito Ariu, Masahiro Kato, Jupei Komiyama, Kenichiro McAlinn, and Chao Qin
    A Comment on "Adaptive Treatment Assignment in Experiments for Policy Choice"
    Revise and Resubmit for Econometrica. [arXiv]
  6. Masahiro Kato, Takuya Ishihara, Junya Honda, and Yusuke Narita
    Efficient Adaptive Experimental Design for Average Treatment Effect Estimation
    First draft: 13 Feb 2020. [arXiv]
    Revise and Resubmit for Journal of the American Statistical Association (JASA).

International Conference Proceedings

  1. Masahiro Kato
    General Bayesian Policy Learning
    In the Conference on Uncertainty in Artificial Intelligence (UAI), 2026.
  2. Masahiro Kato
    ScoreMatchingRiesz: Score Matching for Debiased Machine Learning and Policy Path Estimation
    In the International Conference on Machine Learning (ICML), 2026.
  3. Kiet Q. H. Vo, Siu Lun Chau, Masahiro Kato, Yixin Wang, and Krikamol Muandet
    Strategic Learning with Local Explanations as Feedback
    In the International Conference on Artificial Intelligence and Statistics (AISTATS), 2026. [arXiv]
  4. Masahiro Kato, Fumiaki Kozai, and Ryo Inokuchi
    PUATE: Semiparametric Efficient Average Treatment Effect Estimation from Treated (Positive) and Unlabeled Units
    In the Advances in Neural Information Processing Systems (NeurIPS), 2025. [arXiv]
  5. Yuika Shiina*, Masahiro Kato, and Ryo Inokuchi
    Analysis of the Keiki Watchers Survey using Independent Component Analysis and Linear Discriminant Analysis
    In the 3rd International Conference on Computational and Data Sciences in Economics and Finance (CDEF), 2025.
  6. Masahiro Kato*, Yuki Ikeda, Kentaro Baba, Takashi Imai, and Ryo Inokuchi
    Learning from Double Positive and Unlabeled Data for Potential-Customer Identification
    In the 3rd International Conference on Computational and Data Sciences in Economics and Finance (CDEF), 2025. [arXiv]
  7. Masahiro Kato*
    Neyman Allocation for Two-Armed Gaussian Best-Arm Identification with Unknown Variances
    In the 3rd International Conference on Computational and Data Sciences in Economics and Finance (CDEF), 2025.
  8. Masahiro Kato* and Shinji Ito
    LC-Tsallis-INF: Generalized Best-of-Both-Worlds Linear Contextual Bandits
    In the International Conference on Artificial Intelligence and Statistics (AISTATS), 2025. [openreview]
  9. Masahiro Kato*
    Analysis of the Temporal Structure in Economic Condition Assessments
    In IIAI International Congress on Advanced Applied Informatics (IIAI-AAI), 2024. [IEEE]
  10. Masahiro Kato*, Kentaro Baba, Hibiki Kaibuchi, and Ryo Inokuchi
    Bayesian Portfolio Optimization by Predictive Synthesis
    In IIAI International Congress on Advanced Applied Informatics (IIAI-AAI), 2024. [IEEE]
  11. Masahiro Kato*, Akihiro Oga, Wataru Komatsubara, and Ryo Inokuchi
    Active Adaptive Experimental Design for Treatment Effect Estimation with Covariate Choices
    In the International Conference on Machine Learning (ICML, Oral 1.5%), 2024. [PMLR][slide]
  12. Masahiro Kato*, Masaaki Imaizumi, and Kentaro Minami
    Unified Perspective on Probability Divergence via Maximum Likelihood Density Ratio Estimation: Bridging KL-Divergence and Integral Probability Metrics
    In the International Conference on Artificial Intelligence and Statistics (AISTATS), 2023. [arXiv]
  13. Shota Yasui* and Masahiro Kato* (*Equal contribution)
    Learning Classifiers under Delayed Feedback with a Time Window Assumption
    In the International Conference on Knowledge Discovery and Data Mining (KDD), 2022. [ACM]
  14. Masahiro Kato*, Masaaki Imaizumi, Kenichiro McAlinn, Shota Yasui, and Haruo Kakehi
    Learning Causal Models from Conditional Moment Restrictions by Importance Weighting
    In the International Conference on Learning Representations (ICLR, Spotlight 4% 176/3391), 2022. [openreview][arXiv][slide][poster]
  15. Masahiro Kato*, Kenichiro McAlinn, and Shota Yasui
    The Adaptive Doubly Robust Estimator and a Paradox Concerning Logging Policy
    In the Advances in Neural Information Processing Systems (NeurIPS), 2021. [openreview]
  16. Masahiro Kato* and Takeshi Teshima
    Non-negative Bregman Divergence Minimization for Deep Direct Density Ratio Estimation
    In the International Conference on Machine Learning (ICML), 2021. [PMLR]
  17. Riku Togashi, Masahiro Kato, Mayu Otani, Tetsuya Sakai, and Shin’ichi Satoh
    Scalable Personalised Item Ranking through Parametric Density Estimation
    In the Conference on Research and Development in Information Retrieval (SIGIR), 2021. [ACM]
  18. Riku Togashi, Masahiro Kato, Mayu Otani, and Shin’ichi Satoh
    Density-Ratio Based Personalised Ranking from Implicit Feedback
    In the Web Conference (WWW), 2021. [ACM]
  19. Masatoshi Uehara*, Masahiro Kato*, and Shota Yasui (*Equal contribution)
    Off-Policy Evaluation and Learning for External Validity under a Covariate Shift
    In the Advances in Neural Information Processing Systems (NeurIPS, Spotlight 3% 280/9054), 2020. [NeurIPS]
  20. Masahiro Kato*, Takeshi Teshima, and Junya Honda
    Learning from Positive and Unlabeled Data with a Selection Bias
    In the International Conference on Learning Representations (ICLR), 2019. [openreview]

Journal Articles

  1. Masahiro Kato and Akari Ohda
    Asymptotically Unbiased Synthetic Control Methods by Density Matching
    Journal of Causal Inference. [arXiv]
  2. Masahiro Kato and Kaito Ariu
    The Role of Contextual Information in Best Arm Identification
    Journal of Machine Learning Research (JMLR). [arXiv]
  3. Junpei Komiyama, Kaito Ariu, Masahiro Kato, and Chao Qin
    Optimal Simple Regret in Bayesian Best Arm Identification
    Mathematics of Operations Research.
  4. Masahiro Kato and Shinji Ito
    Best-of-Both-Worlds Linear Contextual Bandits
    Transactions on Machine Learning Research. [openreview]

Domestic Conference Proceedings

  1. Masahiro Kato
    Conformal Predictive Portfolio Selection
    JAFEE 2024 Winter Conference. [arXiv]
  2. 椎名唯圭，加藤真大，井口亮
    独立成分分析とFisherの線形判別による内閣府景気ウォッチャー調査データの分析
    言語処理学会 第31回年次大会. [paper]
  3. 加藤真大，浦川通，田口雄哉，新妻巧朗，田森秀明，羽根田賢和，坂口慶祐，持橋大地
    線形判別分析のPU学習による朝日歌壇短歌の分析
    言語処理学会 第31回年次大会. [paper]
  4. 加藤真大，浦川通，田口雄哉，新妻巧朗，田森秀明，羽根田賢和，持橋大地
    文埋め込みに基づく朝日歌壇短歌の分析
  5. 井口亮，加藤真大，貝淵響，野田俊也，今泉允聡
    密度比マッチングと勾配コミュニケーションによる異質性を伴う連合学習
    第32回 人工知能学会 金融情報学研究会（SIG-FIN）. [paper]
  6. 馬場健太郎，加藤真大，今井岳
    二重PU学習による潜在的顧客の特定
    第32回 人工知能学会 金融情報学研究会（SIG-FIN）. [paper]
  7. 加藤真大，貝淵響
    ベイジアン予測統合に基づくポートフォリオ選択
    第32回 人工知能学会 金融情報学研究会（SIG-FIN）. [paper]

Workshop Presentations

  1. Akira Fukuda, Masahiro Kato, Kenichiro McAlinn, and Kosaku Takanashi
    Bayesian Predictive Synthetic Control Methods
    In ICML 2023 Workshop on Counterfactuals in Minds and Machines. [Google Drive]
  2. Masahiro Kato, Masaaki Imaizumi, Takuya Ishihara, and Toru Kitagawa
    Fixed-Budget Hypothesis Best Arm Identification: On the Information Loss in Experimental Design
    In ICML 2023 Workshop on New Frontiers in Learning, Control, and Dynamical Systems. [openreview]
  3. Masahiro Kato
    Best Arm Identification with a Fixed Budget under a Small Gap
    Allied Social Sciences Association (ASSA) 2023 Annual Meeting. [slide]
  4. Masahiro Kato, Masaaki Imaizumi, Takuya Ishihara, and Toru Kitagawa
    Semiparametric Best Arm Identification with Contextual Information
    In IBIS 2022. [arXiv][poster]
  5. 加藤真大
    経済学と機械学習：因果推論と密度比推定を中心に
    統計・機械学習若手シンポジウム 2022. [slide]
  6. Masahiro Kato
    Recent Findings on Density-Ratio Approaches in Machine Learning
    Workshop on Functional Inference and Machine Intelligence (FIMI) 2022. [slide]
  7. Masahiro Kato, Masaaki Imaizumi, Kenichiro McAlinn, Shota Yasui, and Haruo Kakehi
    Learning Causal Relationships from Conditional Moment Conditions by Importance Weighting
    In NeurIPS 2021 Workshop on Machine Learning meets Econometrics. [arXiv]
  8. Masahiro Kato, Kei Nakagawa, Kenshi Abe, and Tetsuro Morimura
    Direct Expected Quadratic Utility Maximization for Mean-Variance Controlled Reinforcement Learning
    In NeurIPS 2021 Workshop on Deep Reinforcement Learning. [arXiv]
  9. 加藤真大
    効率的な因果推論と意思決定のための実験計画において異質性が果たす役割
    統計関連学会連合大会 2021. [slide]
  10. Masahiro Kato, Shota Yasui, and Kenichiro McAlinn
    The Adaptive Doubly Robust Estimator for Policy Evaluation in Adaptive Experiments
    In ICML 2021 Workshop on The Neglected Assumptions In Causal Inference. [arXiv]
  11. Masahiro Kato, Takuya Ishihara, Junya Honda, and Yusuke Narita
    Adaptive Experimental Design for Efficient Treatment Effect Estimation
    In NeurIPS 2020 Workshop on Causal Discovery and Causality-Inspired Machine Learning.
  12. 加藤真大
    平均処置効果の推定のための適応的実験計画
    慶應計量経済学ワークショップ 2020. [slide]

Other

  1. Masahiro Kato, Kei Nakagawa, Kenshi Abe, Tetsuro Morimura, and Kentaro Baba
    Direct Expected Quadratic Utility Maximization for Mean-Variance Controlled Reinforcement Learning
    Accepted to the IEEE Symposium on Computational Intelligence for Financial Engineering and Economics (CIFEr), without publication in proceedings. [arXiv]
  2. Masahiro Kato
    A Note on Doubly Robust Estimator in Regression Discontinuity Designs
    Technical note. [arXiv]

* denotes lead or co-lead author.