NEURALCSA, a neural framework for causal sensitivity analysis, can learn valid bounds on causal queries under a wide range of sensitivity models, treatment types, and causal queries, including multiple outcomes.
The proposed C-XGBoost model exploits the strong prediction abilities of XGBoost algorithm and the ability of causal inference neural networks to learn representations useful for estimating outcomes in both treatment and control groups, resulting in an effective tree-based ensemble model for causal effect estimation.