Reproduce “Composite Marginal Likelihood Methods for Random Utility Models” 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
Zhibing Zhao, Lirong Xia
RBCML fits ranked preferences using selected position comparisons and weighted alternative pairs in a composite marginal likelihood objective. The authors characterize conditions for consistency in Plackett–Luce and a symmetric random-utility class. On synthetic rankings over 10 alternatives, averaged over 50,000 trials, RBCML has lower scaled error and runtime than GMM in the tested Gaussian setting; Plackett–Luce methods trade off error and runtime.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.