COSTAR: Improved Temporal Counterfactual Estimation with Self-Supervised Learning
This work addresses the challenge of modeling time-dependent confounders in real-world datasets for decision-making, representing an incremental improvement with a novel method for a known bottleneck.
The paper tackles the problem of estimating temporal counterfactual outcomes from observed history, which is crucial for decision-making in domains like healthcare and e-commerce, by introducing COSTAR, a novel approach that integrates self-supervised learning for improved historical representations. It demonstrates superior performance in estimation accuracy and generalization to out-of-distribution data compared to existing models, as validated on synthetic and real-world datasets.
Estimation of temporal counterfactual outcomes from observed history is crucial for decision-making in many domains such as healthcare and e-commerce, particularly when randomized controlled trials (RCTs) suffer from high cost or impracticality. For real-world datasets, modeling time-dependent confounders is challenging due to complex dynamics, long-range dependencies and both past treatments and covariates affecting the future outcomes. In this paper, we introduce Counterfactual Self-Supervised Transformer (COSTAR), a novel approach that integrates self-supervised learning for improved historical representations. We propose a component-wise contrastive loss tailored for temporal treatment outcome observations and explain its effectiveness from the view of unsupervised domain adaptation. COSTAR yields superior performance in estimation accuracy and generalization to out-of-distribution data compared to existing models, as validated by empirical results on both synthetic and real-world datasets.