Statistical Theory and Related Fields

Special Issue on Differential Privacy

Aims & Scope

Differential Privacy (DP) provides rigorous guarantees for privacy-preserving analysis and information sharing. This special issue seeks original research at the interface of statistics, machine learning, and DP, including but not limited to theoretical advances, methodological innovations, empirical studies, applications, and policy perspectives – aiming at trustworthy inference and robust decisions under formal privacy protection. Submissions should be statistically principled, reproducible, and of practical relevance to academia, industry, and the public sector.

Article Types

• Theory & Methods
• Application & Case Studies
• Survey/Position Articles (concise, critical syntheses with clear guidance)
• Software (novel, well-documented tools with examples)

Topics of Interest (include, but are not limited to)

• Statistical theory and methodology under DP, such as estimation, hypothesis testing, confidence intervals, Bayesian inference, regularization, causal inference, survey statistics, and small-area estimation etc.
• Machine Learning and DP, such as differentially private machine learning, deep learning, privacy for foundation models, etc.
• Study design & data collection with DP incorporated.
• Privacy accounting & composition, such as tight accounting, privacy amplification, subsampling, and shuffling.
• Mechanisms & releases: novel DP mechanisms; synthetic data, federated and distributed learning, etc.
• Utility, Evaluation, benchmarking validity: privacy-utility trade-offs,  general utility, inferential utility, learning utility, sensitivity analysis, calibration and coverage,  metrics, standardized tasks, comparative studies, validation/verification methods.
• Robustness & fairness: interactions among privacy, robustness, and equity, and other aspects of trustworthy statistical analysis and machine learning.
• Applications & case studies: official statistics, health, social science, finance/business/economics, mobile computing, etc.
• Systems & software: scalable implementations, software packages,
• Guidelines, Policy & governance: practitioner guidelines and policy and data governance relevant to statistical practice and methodology on data collection and statistical releases with DP guarantees