How does Naïve Bayes behave in federated learning environments?

Naïve Bayes is well-suited for federated learning because it can aggregate sufficient statistics without sharing raw data.

In federated learning, data remains distributed across clients. NB supports this setup by allowing each client to compute local counts and statistics, which are then aggregated centrally. This preserves privacy while enabling global model construction. It is one of the simplest models compatible with federated aggregation.