Long-Term Individual Causal Effect Estimation via Identifiable Latent Representation Learning

2025-05-08

AI Summary

Estimating long-term individual causal effects by integrating longitudinal observational and short-term experimental data remains challenging due to unobserved confounding—yet existing methods rely on strong, often unrealistic assumptions such as no unmeasured confounding or additive confounding bias. This paper proposes a novel framework that does not require such idealized assumptions. Leveraging the natural heterogeneity across multi-source data, we first establish identifiability of latent confounders; then, we develop an end-to-end causal estimation model grounded in identifiable latent representation learning, accompanied by rigorous theoretical identifiability guarantees. Extensive experiments on multiple synthetic and semi-synthetic benchmarks demonstrate that our method significantly outperforms state-of-the-art baselines in both accuracy and robustness. It provides a more reliable and practical solution for long-horizon causal inference under realistic, confounded settings.

Technology Category

• Machine Learning: Causal Learning
• Reasoning under Uncertainty: Causality
• Knowledge Representation and Reasoning: Computational Complexity of Reasoning

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for heterogeneous, signed, attributed, multi-relational, temporal, higher-order, and annotated Web-related graphs
• User Modeling, Personalization and Recommendation: Studies of user behavior, including longitudinal effects of personalized systems

Abstract

Estimating long-term causal effects by combining long-term observational and short-term experimental data is a crucial but challenging problem in many real-world scenarios. In existing methods, several ideal assumptions, e.g. latent unconfoundedness assumption or additive equi-confounding bias assumption, are proposed to address the latent confounder problem raised by the observational data. However, in real-world applications, these assumptions are typically violated which limits their practical effectiveness. In this paper, we tackle the problem of estimating the long-term individual causal effects without the aforementioned assumptions. Specifically, we propose to utilize the natural heterogeneity of data, such as data from multiple sources, to identify latent confounders, thereby significantly avoiding reliance on idealized assumptions. Practically, we devise a latent representation learning-based estimator of long-term causal effects. Theoretically, we establish the identifiability of latent confounders, with which we further achieve long-term effect identification. Extensive experimental studies, conducted on multiple synthetic and semi-synthetic datasets, demonstrate the effectiveness of our proposed method.

Problem

• Estimating long-term individual causal effects without ideal assumptions
• Identifying latent confounders using data heterogeneity from multiple sources
• Developing a latent representation learning-based estimator for causal effects

Innovation

• Utilizes data heterogeneity to identify confounders
• Employs latent representation learning for estimation
• Ensures identifiability of latent confounders theoretically