Privately Answering Queries on Skewed Data via Per-Record Differential Privacy

Summary: Propose per-record zero-concentrated differential privacy (PzCDP), a formalism that makes each record's privacy loss a public function of its confidential value (data-dependent budget). Develop properties and mechanisms for skewed/heavy-tailed statistics to accurately release outlier-driven aggregates while relaxing standard DP noise constraints. (summarized by gpt-5-mini on Feb 09 2026)

Paper ID: 13529
Venue: VLDB
Year: 2024
Pagerank: 4.5435639e-05
Overall Rank: 8,282 | 42.39%
DOI: 10.14778/3681954.3681989

Incoming Non-self Citations Over Time

Authors

Incoming Citations (Sorted by Pagerank)

Showing 1 of 1 citing papers.

Rank	Citing Paper	Year	Venue	Pagerank
10,041	A General Framework for Per-record Differential Privacy	2026	SIGMOD	4.1945683e-05

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

Semantically Similar Papers

Overall Rank	Paper	Year	Venue	Pagerank
1,738	PrivateSQL: A Differentially Private SQL Query Engine	2019	VLDB	0.00010720057
3,760	Output Perturbation with Query Relaxation	2008	VLDB	6.7805033e-05
7,579	A Nearly Instance-optimal Differentially Private Mechanism for Conjunctive Queries	2022	PODS	4.706055e-05
6,970	Architecting a Differentially Private SQL Engine	2019	CIDR	4.8796169e-05
3,097	Publishing Set-Valued Data via Differential Privacy	2011	VLDB	7.5647028e-05
2,434	Optimizing error of high-dimensional statistical queries under differential privacy	2018	VLDB	8.8278955e-05
7,864	Differentially Private Data Release over Multiple Tables	2023	PODS	4.6327272e-05
6,486	Differential Privacy in Data Publication and Analysis	2012	SIGMOD	5.0445043e-05
10,992	Personalized Truncation for Personalized Privacy	2024	SIGMOD	4.1945683e-05
10,041	A General Framework for Per-record Differential Privacy	2026	SIGMOD	4.1945683e-05