Year: 2026
Type: preprint
Abstract: We study the problem of selecting the largest among $n$ unknown values $x_1,\dots,x_n$ given only a single unbiased estimate $y_i$ for each $x_i$. We design strategies that are simultaneously admissible (not uniformly dominated by any other strategy) and also never worse than a given baseline such as uniform random selection. We provide an application to stochastic optimization, where we obtain online-to-batch conversion bounds with a desirable "no-compromise" guarantee: they are never worse than standard random iterate selection, and yet can be significantly better in benign settings.
Source: arXiv (Cornell University)
Institutions: University of Utah, Boston University
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7164576848
Cites:
Cited by:
Related to: