Reproduce “Time Series Structure Learning with Local Causal Parent Discovery” 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
Jacqueline Maasch, Shu Wan, Hank Zhu, Huan Liu, Zhiguang Wang
ITPD learns a separate parent set for each time-indexed variable, rather than assuming causal links recur in stationary windows. It screens candidate parents using time order, causal-graph structure, and conditional-independence tests, then verifies the survivors. On 2,000-node graphs, it ran 2.3× faster than iterative IAMB, 9.1× faster than DF-PC, and 40.9× faster than PCMCI; these are reported benchmarks, not universal speedups.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.