Reproduce “Estimating the normal-inverse-Wishart distribution” 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
Jonathan So
The authors present a procedure for converting valid mean parameters of the normal-inverse-Wishart (NIW) distribution into natural parameters, motivated by expectation propagation. They recover most parameters in closed form and reduce the remaining task, solving for ν, to a one-dimensional root problem. For valid inputs, Newton–Raphson converges to the unique root when initialized with a negative function value, on ν>d−1.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.