activity
20242026
collaborators

9 papers

cs.LG2026

Machine-Learning Emulation of Satellite Greenhouse Gas Retrievals: Stability over Time

Nugzar Gognadze, Motonobu Kanagawa, Yu Someya +1

Retrieval algorithms are used to estimate atmospheric concentrations of greenhouse gases (GHGs), such as carbon dioxide (CO2) and methane (CH4), by solving inverse problems from hi…

cs.LG2026

Predictive Uncertainty in Short-Term PV Forecasting under Missing Data: A Multiple Imputation Approach

Parastoo Pashmchi, Jérôme Benoit, Motonobu Kanagawa

Missing values are common in photovoltaic (PV) power data, yet the uncertainty they induce is not propagated into predictive distributions. We develop a framework that incorporates…

stat.ML2025

kNNSampler: Stochastic Imputations for Recovering Missing Value Distributions

Parastoo Pashmchi, Jérôme Benoit, Motonobu Kanagawa

We study a missing-value imputation method, termed kNNSampler, that imputes a given unit's missing response by randomly sampling from the observed responses of the most similar…

stat.ML2025

Variable Selection in Maximum Mean Discrepancy for Interpretable Distribution Comparison

Kensuke Mitsuzawa, Motonobu Kanagawa, Stefano Bortoli +2

We study two-sample variable selection: identifying variables that discriminate between the distributions of two sets of data vectors. Such variables help scientists understand the…

math.ST2025

Comparing Scale Parameter Estimators for Gaussian Process Interpolation with the Brownian Motion Prior: Leave-One-Out Cross Validation and Maximum Likelihood

Masha Naslidnyk, Motonobu Kanagawa, Toni Karvonen +1

Gaussian process (GP) regression is a Bayesian nonparametric method for regression and interpolation, offering a principled way of quantifying the uncertainties of predicted functi…

stat.ML2025

Gaussian Processes and Reproducing Kernels: Connections and Equivalences

Motonobu Kanagawa, Philipp Hennig, Dino Sejdinovic +1

This monograph studies the relations between two approaches using positive definite kernels: probabilistic methods using Gaussian processes, and non-probabilistic methods using rep…