mlsanityData to 14 Oct 2026

AIM: An Adaptive and Iterative Mechanism for Differentially Private Synthetic Data

arXiv 2201.12677 ↗First posted 29 Jan 2022cs.DBcited by 90

Not on the map. Why?

Abstract (arXiv)

We propose AIM, a new algorithm for differentially private synthetic data generation. AIM is a workload-adaptive algorithm within the paradigm of algorithms that first selects a set of queries, then privately measures those queries, and finally generates synthetic data from the noisy measurements. It uses a set of innovative features to iteratively select the most useful measurements, reflecting both their relevance to the workload and their value in approximating the input data. We also provide analytic expressions to bound per-query error with high probability which can be used to construct confidence intervals and inform users about the accuracy of generated data. We show empirically that AIM consistently outperforms a wide variety of existing mechanisms across a variety of experimental settings.

Key facts. Cited by 90 papers in the map (data through 14 Oct 2026). First posted on arXiv 29 Jan 2022.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2022: 4; 2023: 17; 2024: 21; 2025: 23; 2026: 25.

90 papers.

Sort:

By area:

  1. 2023How to DP-fy ML: A Practical Guide to Machine Learning with Differential Privacycited by 153
  2. 2023Machine Learning for Synthetic Data Generation: A Reviewcited by 120
  3. 2023Differentially Private Diffusion Models Generate Useful Synthetic Imagescited by 86
  4. 2022Private Synthetic Data for Multitask Learning and Marginal Queriescited by 35
  5. 2024Privacy-Preserving Instructions for Aligning Large Language Modelscited by 35
  6. 2023A Linear Reconstruction Approach for Attribute Inference Attacks against Synthetic Datacited by 24
  7. 2023Generating Private Synthetic Data with Genetic Algorithmscited by 18
  8. 2024Systematic Assessment of Tabular Data Synthesiscited by 18
  9. 2024Differentially Private Tabular Data Synthesis using Large Language Modelscited by 16
  10. 2025A Survey on Deep Learning Approaches for Tabular Data Generation: Utility, Alignment, Fidelity, Privacy, Diversity, and Beyondcited by 15
  11. 2024CTSyn: A Foundation Model for Cross Tabular Data Generationcited by 14
  12. 2023Graphical vs. Deep Generative Models: Measuring the Impact of Differentially Private Mechanisms and Budgets on Utilitycited by 13
38 more
  1. 2023CuTS: Customizable Tabular Synthetic Data Generationcited by 13
  2. 2023DP-TBART: A Transformer-based Autoregressive Model for Differentially Private Tabular Data Generationcited by 13
  3. 2024The Elusive Pursuit of Reproducing PATE-GAN: Benchmarking, Auditing, Debuggingcited by 13
  4. 2025Position: All Current Generative Fidelity and Diversity Metrics are Flawedcited by 12
  5. 2025DPImageBench: A Unified Benchmark for Differentially Private Image Synthesiscited by 11
  6. 2025The DCR Delusion: Measuring the Privacy Risk of Synthetic Datacited by 11
  7. 2022Noise-Aware Statistical Inference with Differentially Private Synthetic Datacited by 9
  8. 2023Post-processing Private Synthetic Data for Improving Utility on Selected Measurescited by 9
  9. 2025Synthetic Tabular Data: Methods, Attacks and Defensescited by 9
  10. 2024Privacy-preserving data release leveraging optimal transport and particle gradient descentcited by 8
  11. 2024Joint Selection: Adaptively Incorporating Public Information for Private Synthetic Datacited by 8
  12. 2025From Easy to Hard: Building a Shortcut for Differentially Private Image Synthesiscited by 7
  13. 2025Differential privacy for medical deep learning: methods, tradeoffs, and deployment implicationscited by 7
  14. 2025Struct-Bench: A Benchmark for Differentially Private Structured Text Generationcited by 7
  15. 2023FLAIM: AIM-based Synthetic Data Generation in the Federated Settingcited by 6
  16. 2024Tabular Data Synthesis with Differential Privacy: A Surveycited by 6
  17. 2024Differential Privacy Under Class Imbalance: Methods and Empirical Insightscited by 6
  18. 2023Privately generating tabular data using language modelscited by 5
  19. 2024DP-2Stage: Adapting Language Models as Differentially Private Tabular Data Generatorscited by 5
  20. 2025Is API Access to LLMs Useful for Generating Private Synthetic Tabular Data?cited by 5
  21. 2025The Importance of Being Discrete: Measuring the Impact of Discretization in End-to-End Differentially Private Synthetic Datacited by 5
  22. 2023A Simple and Practical Method for Reducing the Disparate Impact of Differential Privacycited by 4
  23. 2024Does Differentially Private Synthetic Data Lead to Synthetic Discoveries?cited by 4
  24. 2024Private Regression via Data-Dependent Sufficient Statistic Perturbationcited by 4
  25. 2025RelDiff: Relational Data Generative Modeling with Graph-Based Diffusion Modelscited by 4
  26. 2025Improving the Generation and Evaluation of Synthetic Data for Downstream Medical Causal Inferencecited by 4
  27. 2024Bounding the Excess Risk for Linear Models Trained on Marginal-Preserving, Differentially-Private, Synthetic Datacited by 3
  28. 2024Sharper Bounds for Chebyshev Moment Matching, with Applicationscited by 3
  29. 2025dpmm: Differentially Private Marginal Models, a Library for Synthetic Tabular Data Generationcited by 3
  30. 2025ACTG-ARL: Differentially Private Conditional Text Generation with RL-Boosted Controlcited by 3
  31. 2022Private Synthetic Data with Hierarchical Structurecited by 2
  32. 2022Answering Private Linear Queries Adaptively using the Common Mechanismcited by 2
  33. 2023ResidualPlanner+: a scalable matrix mechanism for marginals and beyondcited by 2
  34. 2024A Bias-Variance Decomposition for Ensembles over Multiple Synthetic Datasetscited by 2
  35. 2024Differentially Private Synthetic Data Generation for Relational Databasescited by 2
  36. 2024Privacy without Noisy Gradients: Slicing Mechanism for Generative Model Trainingcited by 2
  37. 2025P2NIA: Privacy-Preserving Non-Iterative Auditingcited by 2
  38. 2025Leveraging Vertical Public-Private Split for Improved Synthetic Data Generationcited by 2

50 most cited of 90 All 90 (JSON)