{"id":"arxiv/2603.20135","url":"https://arxiv.org/abs/2603.20135","tags":["math.ST","cs.IT"],"score":0,"title":"Classifier-Based Nonparametric Sequential Hypothesis Testing","domain":"math.ST","source":"arxiv","authors":["Chia-Yu Hsu","Shubhanshu Shekhar"],"reviews":0,"subject":"math.ST","summary":"We consider the problem of constructing sequential power-one tests where the null and alternative classes are specified indirectly through historical or offline data. More specifically, given an offline dataset consisting of observations from $L+1$ distributions $\\{P_0, P_1, \\ldots, P_L\\}$, and a new unlabeled data stream $\\{X_t: t \\geq 1\\} \\overset{i.i.d}{\\sim} P_θ$, the goal is to decide between the null $H_0: θ= 0$, against the alternative $H_1: θ\\in [L]:=\\{1,\\ldots,L\\}$. Our main methodological contribution is a general approach for designing a level-$α$ power-one test for this problem…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-03-20","reactions":{}}