disclosure risk in sample surveys, k-anonymity, synthetic population, original population, synthetic data,
Federated Learning, decentralized approach to statistical model training, quality of synthetic microdata, synthetic datasets,