Bolt-on Differential Privacy for Scalable Stochastic Gradient Descent-based Analytics

Summary: Bolt-on differential privacy for scalable SGD via output perturbation; seamless integration into Bismarck on a relational DBMS. Novel L2-sensitivity analysis of SGD under limited passes; improved convergence, negligible overhead, and up to 4x test accuracy over prior DP-SGD. (summarized by gpt-5-nano on Feb 09 2026)

Paper ID: h39b465da557f4d0a
Venue: SIGMOD
Year: 2017
Pagerank: 7.0516515e-05
Overall Rank: 3,721 | 75.06%
DOI: 10.1145/3035918.3064047

Incoming Non-self Citations Over Time

Authors

1. Xi Wu (Google)
2. Fengan Li (Google)
3. Arun Kumar (University of California San Diego)
4. Kamalika Chaudhuri (University of California San Diego)
5. Somesh Jha (University of Wisconsin)
6. Jeffrey Naughton (Google)

BibTeX Citation

@inproceedings{wu_sigmod17,
        title = {{Bolt-on Differential Privacy for Scalable Stochastic Gradient Descent-based Analytics}},
        author = {Wu, Xi and Li, Fengan and Kumar, Arun and Chaudhuri, Kamalika and Jha, Somesh and Naughton, Jeffrey},
        series = {{SIGMOD} '17},
        booktitle = {Proceedings of the {ACM} {SIGMOD} International Conference on Management of Data},
        publisher = {Association for Computing Machinery},
        doi = {10.1145/3035918.3064047},
        url = {https://dl.acm.org/doi/10.1145/3035918.3064047},
        year = {2017}
}

Incoming Citations (Sorted by Pagerank)

Showing 6 of 6 citing papers.

Rank	Citing Paper	Year	Venue	Pagerank
3,186	CGM: An Enhanced Mechanism for Streaming Data Collection with Local Differential Privacy	2021	VLDB	7.5392292e-05
5,804	A Neural Database for Differentially Private Spatial Range Queries	2022	VLDB	5.9716248e-05
6,350	A Neural Approach to Spatio-Temporal Data Release with User-Level Differential Privacy	2023	SIGMOD	5.788067e-05
6,603	Tuple-oriented Compression for Large-scale Mini-batch Stochastic Gradient Descent	2019	SIGMOD	5.7204739e-05
10,743	Privacy Loss of Noise Perturbation via Concentration Analysis of A Product Measure	2026	SIGMOD	4.9605343e-05
10,890	PrivSTD: Differentially Private Spatio-temporal Trajectory Density Data Publication	2026	VLDB	4.9605343e-05

Outgoing Citations (Sorted by Pagerank)

Showing 5 of 5 cited papers.

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

Rank	Cited Paper	Year	Venue	Pagerank
105	The MADlib Analytics Library or MAD Skills, the SQL	2012	VLDB	0.00033605348
506	Towards a Unified Architecture for in-RDBMS Analytics	2012	SIGMOD	0.00017158546
2,315	Principled Evaluation of Differentially Private Algorithms using DPBench	2016	SIGMOD	8.6412356e-05
5,307	Functional Mechanism: Regression Analysis under Differential Privacy	2012	VLDB	6.1741542e-05
6,787	PrivGene: Differentially Private Model Fitting Using Genetic Algorithms	2013	SIGMOD	5.6656994e-05

Semantically Similar Papers

#	Overall Rank	Paper	Year	Venue
1	13,146	Privacy in Data Systems	2003	PODS
2	4,676	Private Multiplicative Weights Beyond Linear Queries	2015	PODS
3	1,170	PrivateSQL: A Differentially Private SQL Query Engine	2019	VLDB
4	9,101	Privacy and Accuracy-Aware AI/ML Model Deduplication	2025	SIGMOD
5	8,022	Quantifying identifiability to choose and audit epsilon in differentially private deep learning	2021	VLDB
6	10,615	A General Framework for Per-record Differential Privacy	2026	SIGMOD
7	10,035	DPSUR: Accelerating Differentially Private Stochastic Gradient Descent Using Selective Update and Release	2024	VLDB
8	7,315	Architecting a Differentially Private SQL Engine	2019	CIDR
9	7,682	Private Incremental Regression	2017	PODS
10	6,508	Practical Differentially Private and Byzantine-resilient Federated Learning	2023	SIGMOD