← Back to all news
arXiv Machine Learning By Ruizi Yan, Dennis Frauen, Maresa Schr\"oder, Stefan Feuerriegel

Path-specific harm decomposition: A partial identification framework

Read the original on arXiv Machine Learning →

The paper introduces a path‑specific version of the fraction of negatively affected (FNA) to separate total harm into direct and indirect components in causal mediation settings. Because these components depend on joint distributions of potential outcomes that are not point‑identified, the authors develop a partial identification framework, deriving sharp Makarov bounds and a semiparametric efficient estimator with valid confidence intervals under mild margin conditions. The framework is illustrated through numerical experiments, marking the first study of path‑specific harm decomposition and its orthogonal inference.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

Related stories

arXiv Statistics ML
Aug 25

Model-Agnostic Covariate-Assisted Inference on Partially Identified Causal Effects

The paper introduces a model‑agnostic inference framework for partially identified causal effects that leverages covariate information without requiring discrete covariates or accurate conditional distribution estimates. Using duality theory for optimal transport, the method delivers uniformly valid inference in randomized experiments, is doubly robust in observational settings, achieves asymptotic unbiasedness when nuisance parameters converge semiparametrically, and allows multiplier‑bootstrap selection of covariates and models while remaining computationally efficient. Empirical applications show the approach consistently narrows identified sets and confidence intervals without imposing extra structural assumptions.

By Wenlong Ji, Lihua Lei, Asher Spector
More like this →
arXiv AI
Jul 7

Fixed-Confidence Best-Arm Identification for Causal Mediation Analysis

arXiv:2607. 04315v1 Announce Type: cross Abstract: This paper studies the problem of identifying the treatment that maximizes the expected natural direct potential outcome (NDPO), which captures the potential outcome of an intervention while excluding the pathway transmitted through a mediator that researchers may wish to remove from evaluation.

By Harsh Shrivastava, Yuta Kawakami, Junpei Komiyama, Jin Tian
reinforcement-learning
More like this →