Factor Informed Double Deep Learning For Average Treatment Effect Estimation

2025-08-23

AI Summary

Estimating average treatment effects (ATE) in high-dimensional settings with strongly correlated covariates and sparse, nonlinear effects in both propensity score and outcome models poses significant statistical and computational challenges. Method: This paper proposes FIDDLE—a novel ATE estimation framework integrating factor-augmented sparse neural networks (FAST-NN) with double machine learning. FIDDLE jointly models the response function and propensity score using FAST-NN, enabling adaptive learning of low-dimensional latent structures and automatic variable selection within deep learning, while employing augmented inverse probability weighting for robust estimation. Contribution/Results: Theoretically, FIDDLE achieves model robustness and semiparametric efficiency under mild regularity conditions. Empirically, it substantially outperforms conventional methods—including parametric, tree-based, and standard neural network approaches—on both synthetic and real-world datasets, especially in high-dimensional regimes. FIDDLE delivers superior statistical accuracy while maintaining computational feasibility, offering a practical solution for complex causal inference tasks.

Technology Category

• Machine Learning: Causal Learning
• Reasoning under Uncertainty: Causality
• Search and Optimization: Learning to Search

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: Graph neural networks and deep learning approaches for Web-related graphs
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization

Abstract

We investigate the problem of estimating the average treatment effect (ATE) under a very general setup where the covariates can be high-dimensional, highly correlated, and can have sparse nonlinear effects on the propensity and outcome models. We present the use of a Double Deep Learning strategy for estimation, which involves combining recently developed factor-augmented deep learning-based estimators, FAST-NN, for both the response functions and propensity scores to achieve our goal. By using FAST-NN, our method can select variables that contribute to propensity and outcome models in a completely nonparametric and algorithmic manner and adaptively learn low-dimensional function structures through neural networks. Our proposed novel estimator, FIDDLE (Factor Informed Double Deep Learning Estimator), estimates ATE based on the framework of augmented inverse propensity weighting AIPW with the FAST-NN-based response and propensity estimates. FIDDLE consistently estimates ATE even under model misspecification and is flexible to also allow for low-dimensional covariates. Our method achieves semiparametric efficiency under a very flexible family of propensity and outcome models. We present extensive numerical studies on synthetic and real datasets to support our theoretical guarantees and establish the advantages of our methods over other traditional choices, especially when the data dimension is large.

Problem

• Estimating average treatment effect with high-dimensional correlated covariates
• Addressing model misspecification in causal inference estimation
• Enabling nonparametric variable selection for propensity and outcome models

Innovation

• Factor-augmented deep learning estimators
• Nonparametric variable selection via neural networks
• AIPW framework with FAST-NN based estimates