Reproduce “Sample complexity bounds for the Jensen-Shannon divergence” 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
Oren Richter, Adi Ben-Ari, Tom Talpir, Elad Schneidman
The paper gives sufficient sample-size bounds for distinguishing two distributions with equal prior probabilities using independent samples. It compares a log-likelihood-ratio test, which accumulates each sample’s full evidence, with majority vote over Bayes-optimal single-sample decisions. For JSD d and target error ε, the bounds scale as 1/d and 1/d², respectively; they are guarantees, not minimum sample counts.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.