Reproduce “Finite-Sample and Communication-Efficient Networked Information Aggregation” 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

Mohammadhossein Bateni, Zahra Hadizadeh, MohammadTaghi Hajiaghayi, Mahdi JafariRaviz, Shayan Taherijam

The paper analyzes feature-partitioned linear regression, where learners pass fitted predictions along a path while observing different input columns. It compares row subsampling and Gaussian sketching, which sends k values per edge while retaining all m training examples. Under linear-model assumptions, sketching’s upper bound separates compression error from the m-sample statistical term; this is a theoretical comparison, not an experimental speedup.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.