DBScholar

Back to papers

Projected Federated Averaging with Heterogeneous Differential Privacy

Summary: Introduces PFA for federated learning with heterogeneous differential privacy, projecting private updates onto the top singular subspace from public clients before aggregation. PFA+ enables uploading projected updates, achieving over 99% uplink reduction with preserved utility. (summarized by gpt-5-nano on Feb 09 2026)

Paper ID
13141
Venue
VLDB
Year
2022
Pagerank
6.6990707e-05
Overall Rank
4,446 | 69.50%
DOI
10.14778/3503585.3503592

Incoming Non-self Citations Over Time

Authors

BibTeX Citation

@article{liu_vldb22,
        title = {{Projected Federated Averaging with Heterogeneous Differential Privacy}},
        author = {Liu, Junxu and Lou, Jian and Xiong, Li and Liu, Jinfei and Meng, Xiaofeng},
        journal = {PVLDB},
        series = {{VLDB} '22},
        volume = {15},
        number = {4},
        pages = {828--840},
        doi = {10.14778/3503585.3503592},
        url = {https://doi.org/10.14778/3503585.3503592},
        year = {2022}
}

Incoming Citations (Sorted by Pagerank)

Showing 7 of 7 citing papers.

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