A Partial Initialization Strategy to Mitigate the Overfitting Problem in CATE Estimation with Hidden Confounding
This addresses a critical issue in causal inference for domains like e-commerce and healthcare, but it is incremental as it builds on existing methods with a novel initialization strategy.
The paper tackles the problem of overfitting in conditional average treatment effect (CATE) estimation with hidden confounding by proposing a two-stage pretraining-finetuning framework with partial parameter initialization, validated on two datasets.
Estimating the conditional average treatment effect (CATE) from observational data plays a crucial role in areas such as e-commerce, healthcare, and economics. Existing studies mainly rely on the strong ignorability assumption that there are no hidden confounders, whose existence cannot be tested from observational data and can invalidate any causal conclusion. In contrast, data collected from randomized controlled trials (RCT) do not suffer from confounding but are usually limited by a small sample size. To avoid overfitting caused by the small-scale RCT data, we propose a novel two-stage pretraining-finetuning (TSPF) framework with a partial parameter initialization strategy to estimate the CATE in the presence of hidden confounding. In the first stage, a foundational representation of covariates is trained to estimate counterfactual outcomes through large-scale observational data. In the second stage, we propose to train an augmented representation of the covariates, which is concatenated with the foundational representation obtained in the first stage to adjust for the hidden confounding. Rather than training a separate network from scratch, part of the prediction heads are initialized from the first stage. The superiority of our approach is validated on two datasets with extensive experiments.