Reproduce “A Sharp Mutual Information Bound for Binary Pairs” 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
Qianmian Guo
The authors prove a sharp upper bound on mutual information for every binary pair, in terms of its distribution-dependent KL contraction coefficient, the largest ratio of output to input KL divergence over alternative input laws. The bound is strictly tighter than the generic linear bound for coefficients between zero and one. Equality holds for independent pairs or doubly symmetric binary sources.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.