Research map / Federated learning and privacy / Differential privacy
Privacy accounting and mechanisms
Develop tighter composition bounds and noise mechanisms for differential privacy.
127 accepted papers at ICML, NeurIPS, ICLR, CVPR and AAAI (2016–2026) · an approach in Differential privacy, keyword cluster “estimation · error · algorithms”
State History Tried Checklist Preprints Who Export
Where it stands
Not enough benchmark data Fewer than four papers report the same benchmark, so there is no record to follow.
30 papers in 2023–24 and 42 in 2025–26 (at ICML, NeurIPS, ICLR, CVPR and AAAI); its share of the topic grew, from 14.0% to 15.8%.
How this is decided
A benchmark is tracked when at least four papers of this approach report their own method on the same dataset and metric in at least two different years. Its record is the best number reported up to 2024 compared with the best reported in 2025–26: still improving means at least 1% better, flat means at most 0.2% better with at least two papers from 2025–26 trying. Numbers on a different scale from the rest (a fraction among percentages) and far outliers are left out. Results come from different papers with different setups, so they are signals, not head-to-head comparisons. The share of papers is counted at ICML, NeurIPS, ICLR, CVPR and AAAI.
How it got here
Most cited paper of each year
- 2016 · Privacy Odometers and Filters: Pay-as-you-Go Composition (NeurIPS 2016 · 40 citations)
- 2017 · The Bernstein Mechanism: Function Release under Differential Privacy (AAAI 2017 · 40 citations)
- 2018 · Improving the Gaussian Mechanism for Differential Privacy: Analytical Calibration and Optimal Denoising (ICML 2018 · 77 citations)
- 2019 · Poission Subsampled Rényi Differential Privacy (ICML 2019 · 20 citations)
- 2020 · New Oracle-Efficient Algorithms for Private Synthetic Data Release (ICML 2020 · 22 citations)
- 2021 · The Skellam Mechanism for Differentially Private Federated Learning (NeurIPS 2021 · 34 citations)
- 2022 · Differentially Private Linear Sketches: Efficient Implementations and Applications (NeurIPS 2022 · 11 citations)
- 2023 · Integer Subspace Differential Privacy (AAAI 2023 · 7 citations)
- 2024 · Prior-itizing Privacy: A Bayesian Approach to Setting the Privacy Budget in Differential Privacy (NeurIPS 2024 · 9 citations)
- 2025 · Optimality of Matrix Mechanism on $\ell_p^p$-metric (ICLR 2025 · 1 citation)
- 2026 · Asymptotic Optimality of the High-Dimensional Gaussian Mechanism and Improved Low-Dimensional Mechanisms for Differential Privacy (ICML 2026 · 0 citations)
Earliest papers here
- Privacy Odometers and Filters: Pay-as-you-Go Composition (NeurIPS 2016 · 40 citations)
- The Bernstein Mechanism: Function Release under Differential Privacy (AAAI 2017 · 40 citations)
- Practical Locally Private Heavy Hitters (NeurIPS 2017 · 38 citations)
What has been tried
92 papers here report 551 measured changes (a component added, removed or swapped, compared with a baseline): helped 63%, no clear effect 11%, hurt 17%, mixed 9%. Papers mostly publish changes that helped, so "helped" is over-represented.
Ablations in recent papers
| What was changed | Measured on | Result | Effect | Paper |
|---|---|---|---|---|
| apply PPR to fixed-length chunks of 50 dimensions | DME, d=1000, ε range in the plot · MSE | Hurt | Universal Exact Compression of Differentially Private Mechanisms (NeurIPS 2024 · 3 citations) | |
| use a small number of guesses, allowing abstention | Idealized setting, m = 10^7 canaries · empirical ε | Helped | Auditing $f$-differential privacy in one run (ICML 2025 · 0 citations) | |
| Choose m closer to the expected sampled-set size | DP set representation · space | No clear effect | Differentially Private Set Representations (NeurIPS 2024 · 0 citations) |
What reviewers will expect
Datasets and baselines named in the 28 papers of this approach from 2025–2026 that have a full-text summary. Leaving out the common ones is a frequent review question.
Datasets
- CIFAR-10 · 3 of 28 papers
Baselines
- Laplace mechanism · 3 of 28 papers
- AdaSSP · 2 of 28 papers
Recent preprints
Loading recent arXiv preprints…
Who publishes on it
Institutions on the most papers of this approach since 2023 (of 72 papers; from OpenAlex and the papers' affiliations).
- Google · 17 papers
- Inria · 4 papers
- Massachusetts Institute of Technology · 3 papers
- Carnegie Mellon University · 3 papers
- University of California, Berkeley · 3 papers
- Peking University · 2 papers
- National University of Singapore · 2 papers
- Harvard University · 2 papers
- Craft AI · 2 papers
- University of California, San Diego · 2 papers
Use it
Download BibTeX of 40 key papersOpen the topic in the interactive map
Get alerts. Describe your idea in scime atlas and save it (free scime account): you get a weekly email when new preprints land close to it.
Related approaches
Same keyword cluster
- Private online and user-level learning (116 papers)
- Local-DP and private estimation (113 papers)
- Private data analysis frameworks (105 papers)
- Differentially private clustering (41 papers)
Elsewhere in the topic
- Private training with public data (96 papers)
- Differentially private synthetic data (81 papers)
- DP-SGD refinements (60 papers)
- Differentially private federated learning (35 papers)
Benchmark numbers, ablations and baseline numbers are extracted by AI from each paper's full text and can be wrong; the paper is the authority. Data: results index of 2026-11-02, citation counts from OpenAlex (2026-10-09).