Reproduce “Structure Learning with Adaptive Random Neighborhood Informed MCMC” 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
Alberto Caron, Xitong Liang, Samuel Livingstone, Jim Griffin
PARNI-DAG samples posterior distributions over directed acyclic graphs for observational structure learning, using posterior edge probabilities to guide local proposals and allowing direct edge reversals. In tests with 100 simulated observations on real-world graph structures, it achieved a mean structural Hamming distance of 97.60 ± 2.16 on the 107-node arth150 task, lower than competing methods under roughly equal CPU time.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.