Reproduce “Combining Interventional and Observational Data Using Causal Reductions” 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
Maximilian Ilse, Patrick Forré, Max Welling, Joris M. Mooij
The authors show that multiple latent confounders can be represented by one hidden variable in the treatment’s value space, preserving observational and intervention distributions when the outcome does not cause treatment. This lets observational and interventional data share a model. In synthetic flow experiments, 13 of 15 datasets showed substantial reductions in interventional samples needed, but gains were not universal.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.