Reproduce “Generalized Coverage for More Robust Low-Budget Active Learning” 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

Wonho Bae, Junhyug Noh, Danica J. Sutherland

MaxHerding selects images from an unlabeled pool by maximizing generalized coverage: each example contributes its best Gaussian similarity to selected examples. Unlike ProbCover’s fixed-radius test, this avoids repeatedly crediting overlapping selections. In 1-NN tests, it generally leads four benchmarks while ProbCover leads on TinyImageNet; these results cover the reported experiments.

Summary by alphaXiv.

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