pith. sign in

arxiv: 2606.12763 · v1 · pith:FRBRO3XWnew · submitted 2026-06-11 · 💻 cs.LG · cs.DS

Adaptive Weighted Averaging

classification 💻 cs.LG cs.DS
keywords givenneverrandomselectionworseadaptiveadmissibleapplication
0
0 comments X
read the original 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.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.