{"id":"arxiv/2604.07153","url":"https://arxiv.org/abs/2604.07153","tags":["math.ST","stat.ME"],"score":0,"title":"Non-asymptotic two-sample kernel testing with the spectrally truncated normalized MMD","domain":"math.ST","source":"arxiv","authors":["Perrine Lacroix","Bertrand Michel","Franck Picard","Vincent Rivoirard"],"reviews":0,"subject":"math.ST","summary":"Kernel methods provide a flexible and powerful framework for nonparametric statistical testing by embedding probability distributions into a reproducing kernel Hilbert space (RKHS). In this work, we study the kernel two-sample testing problem and focus on a normalized version of the Maximum Mean Discrepancy (MMD) as a test statistic, which scales the discrepancy by the within-group covariance operator to account for data variability. This normalization has been shown to improve test power in both theoretical and empirical settings. Because this normalization requires regularization, we study…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-04-08","reactions":{}}