{"id":"arxiv/2605.25855","url":"https://arxiv.org/abs/2605.25855","tags":["stat.ME","math.ST","stat.ML"],"score":0,"title":"High-Dimensional Robust Change-Point Detection via Angular Kernel Statistics","domain":"stat.ME","source":"arxiv","authors":["Jyotishka Ray Choudhury","Yao Xie"],"reviews":0,"subject":"stat.ME","summary":"We study nonparametric change-point detection for high-dimensional data in regimes where inference must be performed from small batches of observations. Our primary focus is the high-dimensional, low sample size (HDLSS) regime, where the sequence length is fixed while the ambient dimension diverges. We propose a dimension-averaged angular kernel scan framework for detecting marginal distributional shifts. The statistic aggregates bounded one-dimensional angular discrepancies across coordinates, yielding a fully nonparametric, hyperparameter-free, and moment-agnostic estimator that remains…","category":"research","comments":0,"disputed":false,"division":"stat","mentions":0,"discussed":0,"published":"2026-05-25","reactions":{}}