Reproduce “A Single-Sample Polylogarithmic Regret Bound for Nonstationary Online Linear Programming” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Haoran Xu, Owen Shen, Peter Glynn, Yinyu Ye, Patrick Jaillet
The algorithm allocates online resources by repricing remaining capacity before each request, using current resources and one advance sample for each unarrived order. It accepts a request only when its reward exceeds the price of its resource use and capacity remains. Under stated regularity assumptions, its expected regret against the fractional offline linear-programming benchmark is O((log n)^2), not a universal guarantee.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.