There's an old saying that applies well to artificial intelligence and the data that powers it: "Garbage in, garbage out." Gartner found that only 47% of ML/AI models go from prototype to production. These models are complex, with many elements affecting their success. For instance, if you create models to expand your market share, they need to be flexible to adapt to the many external market factors. All this to say that you need to keep in mind that when it comes to AI/ML models, one size does not fit all.

Federated learning is an appealing framework for analyzing sensitive data from distributed health data networks due to its protection of data privacy. Under this framework, data partners at local sites collaboratively build an analytical model under the orchestration of a coordinating site, while keeping the data decentralized. However, existing federated learning methods mainly assume data across sites are homogeneous samples of the global population, hence failing to properly account for the extra variability across sites in estimation and inference. Drawing on a multi-hospital electronic health records network, we develop an efficient and interpretable tree-based ensemble of personalized treatment effect estimators to join results across hospital sites, while actively modeling for the heterogeneity in data sources through site partitioning. The efficiency of our method is demonstrated by a study of causal effects of oxygen saturation on hospital mortality and backed up by comprehensive numerical results.