Reproduce “Intervention Generalization: A View from Factor Graph 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
Gecia Bravo-Hermsdorff, David S. Watson, Jialin Yu, Jakob Zeitler, Ricardo Silva
The authors introduce an interventional factor model (IFM) representing system distributions as local factors with intervention dependencies specified in advance. Under factorization and support conditions, observed regimes can identify some untested ones. In a semi-synthetic Sachs evaluation, IFM2 achieves pRMSE 0.111, versus 0.414 for a black-box model and 0.408 for a DAG model. This is semi-synthetic evidence, not real-world validation.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.