Causal EpiNets: Precision-corrected Bounds on Individual Treatment Effects using Epistemic Neural Networks
Gandharv Patil, Keyi Tang, Raquel Aoki, Leo Guelman
TLDR
Causal EpiNets use an anchored neural architecture and Epistemic Neural Networks to provide precision-corrected, valid bounds on individual treatment effects.
Key contributions
- Introduces Causal EpiNets, a neural framework for robustly estimating individual treatment effect bounds (PNS).
- Uses an anchored neural architecture to guarantee satisfaction of structural probability constraints.
- Leverages Epistemic Neural Networks for precision-corrected uncertainty quantification, resolving extremum bias.
- Achieves nominal coverage and valid constraints in high-dimensional data, outperforming standard methods.
Why it matters
Individual treatment effects are vital for personalized decisions. Current methods for bounding these effects often fail in finite samples, yielding unreliable results. This paper introduces a robust neural framework that ensures valid and accurate bounds, significantly advancing causal inference.
Original Abstract
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.
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.