Skellam Mixture Mechanism: a Novel Approach to Federated Learning with Differential Privacy
Summary: Skellam Mixture Mechanism (SMM) for DP in federated learning with MPC reduces noise by using real-valued gradients via Skellam mixtures, while keeping updates confidential. Eliminates integer-gradient requirements, delivering stronger DP utility and robust empirical gains. (summarized by gpt-5-nano on Feb 09 2026)
Incoming Non-self Citations Over Time
Authors
- 1. Ergute Bao (National University of Singapore)
- 2. Yizheng Zhu (National University of Singapore)
- 3. Xiaokui Xiao (National University of Singapore)
- 4. Yin Yang (Hamad Bin Khalifa University)
- 5. Beng Chin Ooi (National University of Singapore)
- 6. Benjamin Hong Meng Tan (Agency for Science, Technology and Research)
- 7. Khin Mi Mi Aung (Agency for Science, Technology and Research)
BibTeX Citation
@article{bao_vldb22,
title = {{Skellam Mixture Mechanism: a Novel Approach to Federated Learning with Differential Privacy}},
author = {Bao, Ergute and Zhu, Yizheng and Xiao, Xiaokui and Yang, Yin and Ooi, Beng Chin and Tan, Benjamin Hong Meng and Aung, Khin Mi Mi},
journal = {PVLDB},
series = {{VLDB} '22},
volume = {15},
number = {11},
pages = {2348--2360},
doi = {10.14778/3551793.3551798},
url = {https://doi.org/10.14778/3551793.3551798},
year = {2022}
}Incoming Citations (Sorted by Pagerank)
Showing 5 of 5 citing papers.
| Rank | Citing Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 6,376 | DProvDB: Differentially Private Query Processing with Multi-Analyst Provenance | 2023 | SIGMOD | 5.893174e-05 |
| 7,911 | Incentive-Aware Decentralized Data Collaboration | 2023 | SIGMOD | 5.5181056e-05 |
| 8,456 | Secure and Verifiable Data Collaboration with Low-Cost Zero-Knowledge Proofs | 2024 | VLDB | 5.4217837e-05 |
| 10,914 | Calibrating Noise for Group Privacy in Subsampled Mechanisms | 2025 | VLDB | 5.093636e-05 |
| 11,411 | FedCSS: Joint Client-and-Sample Selection for Hard Sample-Aware Noise-Robust Federated Learning | 2023 | SIGMOD | 5.093636e-05 |
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 0 of 0 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
| Rank | Cited Paper | Year | Venue | Pagerank |
|---|
Previous
Page 1 / 1
Next
Semantically Similar Papers
| # | Overall Rank | Paper | Year | Venue |
|---|---|---|---|---|
| 1 | 7,517 | Private Incremental Regression | 2017 | PODS |
| 2 | 7,833 | An Introduction to Federated Computation | 2022 | SIGMOD |
| 3 | 6,353 | Differentially Private Vertical Federated Clustering | 2023 | VLDB |
| 4 | 5,588 | Federated Matrix Factorization with Privacy Guarantee | 2022 | VLDB |
| 5 | 4,106 | Secure Shapley Value for Cross-Silo Federated Learning | 2023 | VLDB |
| 6 | 6,366 | Practical Differentially Private and Byzantine-resilient Federated Learning | 2023 | SIGMOD |
| 7 | 4,446 | Projected Federated Averaging with Heterogeneous Differential Privacy | 2022 | VLDB |
| 8 | 11,252 | Uldp-FL: Federated Learning with Across-Silo User-Level Differential Privacy | 2024 | VLDB |
| 9 | 7,841 | Quantifying identifiability to choose and audit epsilon in differentially private deep learning | 2021 | VLDB |
| 10 | 10,914 | Calibrating Noise for Group Privacy in Subsampled Mechanisms | 2025 | VLDB |