Back to AI News
arxiv

Probabilities of Causation in Nondeterministic Causal Models

This paper extends the study of probabilities of causation to nondeterministic causal models, where mechanisms remain stochastic even after conditioning on latent variables. The authors propose a generic identification program using moment constraints and constrained optimization to compute sharp bounds on counterfactual quantities. They derive analytic bounds under various assumptions and show these are often more informative than deterministic alternatives, with applications to personalized treatment decisions.

AI Brief

Provides a novel theoretical framework and sharper bounds for causal inference in nondeterministic settings, directly useful for researchers and practitioners in causal AI.

arXiv:2611.18245v1 Announce Type: new Abstract: We study probabilities of causation in nondeterministic causal models, where mechanisms may remain intrinsically stochastic even after all latent variables are fixed. Focusing on binary treatments and outcomes, we develop a generic identification program that represents observational and interventional data as moment constraints on latent stochastic mechanism types and computes minimal/maximal values of counterfactual quantities via constrained optimization. Within this framework, we derive sharp analytic bounds on the probabilities of causation under several assumptions tailored to nondeterministic models. Resulting identification regions are often more informative than those attainable under deterministic alternatives. We illustrate the practical relevance of these results in a case study on personalized treatment decisions and experiments on semi-synthetic datasets. Our results clarify how classical bounds relate to nondeterministic causal semantics, generalizing previous works and providing a unified route from qualitative structural restrictions to computable partial identification regions.
Be the first to rate this article.

Comments & Questions (0)

Captcha
Please be respectful — let's keep the conversation friendly.

No comments yet

Be the first to comment!