Reproduce “The Causal Marginal Polytope for Bounding Treatment Effects” 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
Jakob Zeitler, Ricardo Silva
The causal marginal polytope method bounds causal quantities under hidden confounding by fitting overlapping causal models and requiring shared distributions to agree. This local-consistency relaxation can incorporate intervention data and elicited knowledge, but need not correspond to a realizable model. In a four-variable simulation, overlap raised lower bounds on intervention probabilities from 0.3 to 0.4 and from 0.3 to 0.6.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.