Uldp-FL: Federated Learning with Across-Silo User-Level Differential Privacy
Summary: Uldp-FL: first cross-silo FL framework to guarantee user-level DP when users' records span multiple silos via per-user weighted clipping instead of group-privacy. Provides formal privacy/utility analysis, a private weighting protocol, and improved DP–utility trade-offs. (summarized by gpt-5-mini on Feb 09 2026)
Incoming Non-self Citations Over Time
No non-self incoming citations found for this paper in this database.
Authors
- 1. Fumiyuki Kato
- 2. Li Xiong
- 3. Shun Takagi
- 4. Yang Cao
- 5. Masatoshi Yoshikawa
Incoming Citations (Sorted by Pagerank)
Showing 0 of 0 citing papers.
| Rank | Citing Paper | Year | Venue | Pagerank |
|---|
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 3 of 3 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
| Rank | Cited Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 83 | Privacy Integrated Queries: An Extensible Platform for Privacy-Preserving Data Analysis | 2009 | SIGMOD | 0.00053933811 |
| 8,512 | Network Shuffling: Privacy Amplification via Random Walks | 2022 | SIGMOD | 4.4947966e-05 |
| 11,238 | Olive: Oblivious Federated Learning on Trusted Execution Environment Against the Risk of Sparsification | 2023 | VLDB | 4.1945683e-05 |
Previous
Page 1 / 1
Next