PITE: Multi-Prototype Alignment for Individual Treatment Effect Estimation

TL;DR

This paper introduces PITE, a novel multi-prototype alignment method that captures local group structures and aligns prototypes across treatment groups to improve the accuracy and robustness of individual treatment effect estimation from observational data.

Contribution

PITE is the first end-to-end method that combines local prototype-based clustering with cross-group alignment for better ITE estimation.

Findings

PITE outperforms 13 state-of-the-art methods on benchmark datasets.

It achieves more accurate ITE estimates by preserving local structures.

The method effectively reduces distribution shift through prototype-level alignment.

Abstract

Estimating Individual Treatment Effects (ITE) from observational data is challenging due to confounding bias. Most studies tackle this bias by balancing distributions globally, but ignore individual heterogeneity and fail to capture the local structure that represents the natural clustering among individuals, which ultimately compromises ITE estimation. While instance-level alignment methods consider heterogeneity, they similarly overlook the local structure information. To address these issues, we propose an end-to-end Multi-\textbf{P}rototype alignment method for \textbf{ITE} estimation (\textbf{PITE}). PITE effectively captures local structure within groups and enforces cross-group alignment, thereby achieving robust ITE estimation. Specifically, we first define prototypes as cluster centroids based on similar individuals under the same treatment. To identify local similarity and the distribution consistency, we perform instance-to-prototype matching to assign individuals to the nearest prototype within groups, and design a multi-prototype alignment strategy to encourage the matched prototypes to be close across treatment arms in the latent space. PITE not only reduces distribution shift through fine-grained, prototype-level alignment, but also preserves the local structures of treated and control groups, which provides meaningful constraints for ITE estimation. Extensive evaluations on benchmark datasets demonstrate that PITE outperforms 13 state-of-the-art methods, achieving more accurate and robust ITE estimation.