Reproduce “Logic-Logit: A Logic-Based Approach to Choice Modeling” 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
Shuhan Zhang, Wendi Ren, Shuang Li
Logic-Logit models option features as binary predicates, combines them into weighted AND rules, and uses a mixture logit choice model. Frank–Wolfe and column generation search large conjunction spaces. In Expedia data, it represents uneven booking patterns across hotel ranks, without showing rank caused them. On processed MIMIC-IV treatment-choice data, its 50-rule model achieves top-1 accuracy of 0.264 ± 0.029, versus RUMNet’s 0.1000 ± 0.064, averaged over 10 runs.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.