DPSUR: Accelerating Differentially Private Stochastic Gradient Descent Using Selective Update and Release

Summary: DPSUR speeds DPSGD by privately validating each gradient and applying only updates that steer convergence, cutting noise-induced variance. Privacy preserved via randomized clipping and thresholded selection/release, improving convergence and utility. (summarized by gpt-5-mini on Feb 09 2026)

Paper ID: 13368
Venue: VLDB
Year: 2024
Pagerank: 4.299267e-05
Overall Rank: 9,708 | 32.47%
DOI: 10.14778/3648160.3648164

Incoming Non-self Citations Over Time

Authors

Incoming Citations (Sorted by Pagerank)

Showing 2 of 2 citing papers.

Rank	Citing Paper	Year	Venue	Pagerank
4,115	Federated Heavy Hitter Analytics with Local Differential Privacy	2025	SIGMOD	6.4381114e-05
10,499	Privacy and Accuracy-Aware AI/ML Model Deduplication	2025	SIGMOD	4.1945683e-05

Outgoing Citations (Sorted by Pagerank)

Showing 1 of 1 cited papers.

Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.

Rank	Cited Paper	Year	Venue	Pagerank
11,244	Trajectory Data Collection with Local Differential Privacy	2023	VLDB	4.1945683e-05

Semantically Similar Papers

Overall Rank	Paper	Year	Venue	Pagerank
13,203	High Dimensional Differentially Private Stochastic Optimization with Heavy-tailed Data	2022	PODS	-
10,195	Privacy Loss of Noise Perturbation via Concentration Analysis of A Product Measure	2026	SIGMOD	4.1945683e-05
9,497	DP-Sync: Hiding Update Patterns in Secure Outsourced Databases with Differential Privacy	2021	SIGMOD	4.3341665e-05
5,669	Skellam Mixture Mechanism: a Novel Approach to Federated Learning with Differential Privacy	2022	VLDB	5.380575e-05
7,797	Quantifying identifiability to choose and audit epsilon in differentially private deep learning	2021	VLDB	4.6482625e-05
10,041	A General Framework for Per-record Differential Privacy	2026	SIGMOD	4.1945683e-05
6,459	Practical Differentially Private and Byzantine-resilient Federated Learning	2023	SIGMOD	5.0556005e-05
6,914	Private Incremental Regression	2017	PODS	4.8925595e-05
3,638	Bolt-on Differential Privacy for Scalable Stochastic Gradient Descent-based Analytics	2017	SIGMOD	6.8952488e-05
10,499	Privacy and Accuracy-Aware AI/ML Model Deduplication	2025	SIGMOD	4.1945683e-05