Contrastive representations of high-dimensional, structured treatments
This addresses a key limitation in causal inference for real-world applications where treatments are complex, such as in healthcare or policy, though it builds incrementally on existing representation learning approaches.
The paper tackles the challenge of estimating causal effects when treatments are high-dimensional structured objects like text or video, showing that naive use of shared structure can introduce bias. It proposes a contrastive representation learning method that identifies causal factors and discurs non-causal ones, proving unbiased estimation and validating results on synthetic and real-world datasets with empirical benchmarks.
Estimating causal effects is vital for decision making. In standard causal effect estimation, treatments are usually binary- or continuous-valued. However, in many important real-world settings, treatments can be structured, high-dimensional objects, such as text, video, or audio. This provides a challenge to traditional causal effect estimation. While leveraging the shared structure across different treatments can help generalize to unseen treatments at test time, we show in this paper that using such structure blindly can lead to biased causal effect estimation. We address this challenge by devising a novel contrastive approach to learn a representation of the high-dimensional treatments, and prove that it identifies underlying causal factors and discards non-causally relevant factors. We prove that this treatment representation leads to unbiased estimates of the causal effect, and empirically validate and benchmark our results on synthetic and real-world datasets.