Ensembled Prediction Intervals for Causal Outcomes Under Hidden Confounding
This work addresses causal inference under hidden confounding for researchers, offering an incremental improvement in interval tightness using existing sensitivity models.
The paper tackles the problem of causal inference with hidden confounders by introducing Caus-Modens, an approach that uses deep ensembles to tighten prediction intervals for causal outcomes, achieving smaller interval sizes while maintaining coverage in benchmarks.
Causal inference of exact individual treatment outcomes in the presence of hidden confounders is rarely possible. Recent work has extended prediction intervals with finite-sample guarantees to partially identifiable causal outcomes, by means of a sensitivity model for hidden confounding. In deep learning, predictors can exploit their inductive biases for better generalization out of sample. We argue that the structure inherent to a deep ensemble should inform a tighter partial identification of the causal outcomes that they predict. We therefore introduce an approach termed Caus-Modens, for characterizing causal outcome intervals by modulated ensembles. We present a simple approach to partial identification using existing causal sensitivity models and show empirically that Caus-Modens gives tighter outcome intervals, as measured by the necessary interval size to achieve sufficient coverage. The last of our three diverse benchmarks is a novel usage of GPT-4 for observational experiments with unknown but probeable ground truth.