Causal EpiNets: Precision-corrected Bounds on Individual Treatment Effects using Epistemic Neural Networks

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Patil, Gandharv, Tang, Keyi, Aoki, Raquel, Guelman, Leo
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866914542876884992
author Patil, Gandharv
Tang, Keyi
Aoki, Raquel
Guelman, Leo
author_facet Patil, Gandharv
Tang, Keyi
Aoki, Raquel
Guelman, Leo
contents Individual treatment effects are not point-identified from data. The Probability of Necessity and Sufficiency (PNS) circumvents this limitation by characterizing individual-level causality through intersection bounds derived from combined experimental and observational data. In finite samples, however, standard plug-in estimators systematically fail: they violate structural probability constraints and suffer from extremum bias induced by max-min operators, yielding spuriously narrow intervals. We propose a neural framework for finite-sample PNS estimation that resolves both pathologies. We introduce an anchored neural architecture that guarantees structural constraint satisfaction by construction. To correct extremum bias, we employ precision-corrected intersection-bound inference, leveraging Epistemic Neural Networks for scalable, high-dimensional uncertainty quantification. Empirical evaluations confirm that this approach maintains nominal coverage and exact constraint validity in high-dimensional regimes where standard estimators systematically undercover.
format Preprint
id arxiv_https___arxiv_org_abs_2605_07065
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Causal EpiNets: Precision-corrected Bounds on Individual Treatment Effects using Epistemic Neural Networks
Patil, Gandharv
Tang, Keyi
Aoki, Raquel
Guelman, Leo
Machine Learning
Artificial Intelligence
Econometrics
Individual treatment effects are not point-identified from data. The Probability of Necessity and Sufficiency (PNS) circumvents this limitation by characterizing individual-level causality through intersection bounds derived from combined experimental and observational data. In finite samples, however, standard plug-in estimators systematically fail: they violate structural probability constraints and suffer from extremum bias induced by max-min operators, yielding spuriously narrow intervals. We propose a neural framework for finite-sample PNS estimation that resolves both pathologies. We introduce an anchored neural architecture that guarantees structural constraint satisfaction by construction. To correct extremum bias, we employ precision-corrected intersection-bound inference, leveraging Epistemic Neural Networks for scalable, high-dimensional uncertainty quantification. Empirical evaluations confirm that this approach maintains nominal coverage and exact constraint validity in high-dimensional regimes where standard estimators systematically undercover.
title Causal EpiNets: Precision-corrected Bounds on Individual Treatment Effects using Epistemic Neural Networks
topic Machine Learning
Artificial Intelligence
Econometrics
url https://arxiv.org/abs/2605.07065