Reproduce “Algorithmic recourse under imperfect causal knowledge: a probabilistic approach” 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

Amir-Hossein Karimi, Julius von Kügelgen, Bernhard Schölkopf, Isabel Valera

The paper develops recourse methods for uncertain causal effects when equations are unknown. A Gaussian-process approach models additive Gaussian noise for individual counterfactuals; a CATE/CVAE approach estimates intervention effects for subpopulations. In a seven-variable semi-synthetic loan experiment, GP achieved 100% validity versus 19% for a linear point-estimate method, but at higher cost; both require a causal graph and causal sufficiency.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.