Decision-Aware Proximal Bridge Learning for Optimal Treatment Selection

TL;DR

This paper introduces a decision-aware weighted bridge loss for proximal causal inference, improving treatment selection by focusing on decision-relevant regions and demonstrating reduced regret in empirical tests.

Contribution

It proposes a novel weighted bridge loss tailored for decision-making in proximal causal inference, bridging the gap between treatment effect estimation and optimal treatment selection.

Findings

Decision-aware weighting reduces treatment-selection regret.

The framework improves treatment choice accuracy in proximal causal inference.

Empirical results show enhanced performance across several bridge solvers.

Abstract

Individualized treatment selection with continuous actions requires accurate causal response estimation in decision-relevant regions, rather than uniformly over the entire action space. Estimating a global causal response surface and then choosing the treatment that maximizes it can therefore be suboptimal, since standard estimation objectives allocate modeling effort according to the observed treatment distribution rather than the regions that determine the optimal decision. While decision-aware approaches have been studied in unconfounded settings, this problem remains underexplored in proximal causal inference, where proxy variables and bridge functions enable identification under suitable assumptions even in the presence of hidden confounding. Despite recent progress, proximal methods have primarily focused on treatment-effect and potential-outcome estimation rather than treatment selection and optimal decision-making. To bridge this gap, we introduce a policy-targeted weighted bridge loss that emphasizes decision-relevant treatment regions while retaining global stabilization. We prove a regret bound showing that the proposed weighted bridge loss controls treatment-selection regret through a weighted ill-posedness constant. We instantiate the framework in decision-aware variants of several proximal bridge solvers, yielding practical algorithms that alternate between weighted bridge estimation, response-surface projection, policy update, and weight refinement. Empirically, we find that decision-aware weighting reduces regret across several bridge solvers, suggesting improved treatment selection in proximal settings.