About Me

I’m a research scientist at CyberAgent, Inc.

My research interest is the intersection of Machine Learning and Economics(not only Causal Inference!). Especially, I am interested in the sampling/selection bias in the real-world application. Also, I am interested in how we can combine Mechanism Design, Causal Inference, and Machine Learning.

Research Topics

  • Counterfactual Machine Learning
    • Off-Policy Evaluation/Learning
  • Machine Learning for Causal Inference
    • ITE Prediction
    • Uplift Modeling
  • Computational Advertising
    • Delayed Feedback
    • Advertising Auction

Recent Update

  • 2023.09.29: Our paper “Distributional Treatment Effects of Content Promotion: Empirical Evidence from an ABEMA Field Experiment” has been accepted to CODE@MIT’23!
  • 2023.05.09: We translated “Causal Inference: The Mixtape” into Japanese! amazon.jp link
  • 2022.05.19: Our paper “Delayed Feedback Modeling with a Time Window Assumption” has been accepted to KDD’22!
  • 2021.07.07: Our paper “The Adaptive Doubly Robust Estimator for Policy Evaluation in Adaptive Experiments and a Paradox Concerning Logging Policy” has been accepted to NeurIPS’21.
  • 2021.07.07: Our paper “Debiased Off-Policy Evaluation for Recommendation Systems” has been accepted to RecSys’21.
  • 2020.09.26: Our paper “Off-Policy Evaluation and Learning for External Validity under a Covariate Shift” has been accepted to NeurIPS’20.
  • 2020.06.01: Our paper “Counterfactual Cross-Validation: Stable Model Selection Procedure for Causal Inference Models” has been accepted to ICML’20.
  • 2020.04.23: Our paper “ Dual Learning Algorithm for Delayed Conversions” has been accepted to SIGIR’20.
  • 2020.1.18: I wrote an introductory level Causal Inference/Econometrics text book and now it is on sale! amazon.jp link
  • 2020.1.10: Our paper “A Feedback Shift Correction in Predicting Conversion Rates under Delayed Feedback” has been accepted to The web conference 2020!
  • 2019.10.21: Our papers “Dual Learning Algorithm for Delayed Feedback in Display Advertising” and “Unbiased Pairwise Learning from Implicit Feedback” have been accepted to CausalML Workshop at NeurIPS’19.
  • 2019.8.20: Our paper “Counterfactual Cross Validation” has been accepted to REVEAL Workshop at RecSys’19.
  • 2019.8.20: Our paper “Reinforcement Learning meets Double Machine Learning” has been accepted to REVEAL Workshop at RecSys’19.
  • 2018.12.4: Our paper “Efficient Counterfactual Learning from Bandit Feedback” has been accepted to AAAI 2019!