Reproduce “Non-Adaptive 1-Bit Mean Estimation: Minimax Rates and the Sample-Interval Tradeoff” 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
Ivan Lau, Jonathan Scarlett
The authors show that one-bit mean estimation matches adaptive minimax sample rates without query-side interaction. Their protocol fixes localization and refinement queries in advance, collects responses in parallel, and uses a decoded center to interpret refinement bits. For every fixed k>1, the guarantee covers means in [-λ, λ] with kth central moment at most σ^k. Restricting queries to s intervals adds order λσ/(sε²) log(1/δ) samples.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.