معرفی
Motonobu Kanagawa is an Assistant Professor (Maitre de Conférences) in the Data Science Department at EURECOM in France since 2019. Previously, he held research positions at the University of Tübingen and the Max Planck Institute for Intelligent Systems in Germany under Prof. Philipp Hennig. He earned his PhD in 2016 from the Institute of Statistical Mathematics in Tokyo under Prof. Kenji Fukumizu.
His research focuses on statistical methodologies for complex systems simulation, including reliability validation of simulators, Gaussian processes, Bayesian optimization, and kernel methods. Recent work explores variable selection in distribution comparison and covariate shift adaptation.
Key achievements include organizing the inaugural ProbNum 2025 conference, receiving the Chris Daykin Prize (2023) for pension system analysis with Bayesian optimization, and securing a chair position at the 3IA Côte d'Azur AI institute (2021). He reviews for leading journals like JMLR and conferences such as NeurIPS and ICML.
Publications span topics like k-NN regression optimization, Gaussian process scale parameter estimation, and counterfactual mean embeddings. His work bridges theoretical statistics and applied machine learning, emphasizing computational efficiency and uncertainty quantification.
