Convex optimization for linear query processing under approximate differential privacy

Ganzhao Yuan, Yin Yang, Zhenjie Zhang, Zhifeng Hao

Research output: Chapter in Book/Report/Conference proceedingConference contribution

5 Citations (Scopus)

Abstract

Differential privacy enables organizations to collect accurate aggregates over sensitive data with strong, rigorous guarantees on individuals' privacy. Previous work has found that under differential privacy, computing multiple correlated aggregates as a batch, using an appropriate strategy, may yield higher accuracy than computing each of them independently. However, finding the best strategy that maximizes result accuracy is non-trivial, as it involves solving a complex constrained optimization program that appears to be non-convex. Hence, in the past much effort has been devoted in solving this non-convex optimization program. Existing approaches include various sophisticated heuristics and expensive numerical solutions. None of them, however, guarantees to find the optimal solution of this optimization problem. This paper points out that under (ϵ, δ)-differential privacy, the optimal solution of the above constrained optimization problem in search of a suitable strategy can be found, rather surprisingly, by solving a simple and elegant convex optimization program. Then, we propose an efficient algorithm based on Newton's method, which we prove to always converge to the optimal solution with linear global convergence rate and quadratic local convergence rate. Empirical evaluations demonstrate the accuracy and efficiency of the proposed solution.

Original languageEnglish
Title of host publicationKDD 2016 - Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining
PublisherAssociation for Computing Machinery
Pages2005-2014
Number of pages10
Volume13-17-August-2016
ISBN (Electronic)9781450342322
DOIs
Publication statusPublished - 13 Aug 2016
Externally publishedYes
Event22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2016 - San Francisco, United States
Duration: 13 Aug 201617 Aug 2016

Other

Other22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2016
CountryUnited States
CitySan Francisco
Period13/8/1617/8/16

Fingerprint

Query processing
Convex optimization
Constrained optimization
Newton-Raphson method

ASJC Scopus subject areas

  • Software
  • Information Systems

Cite this

Yuan, G., Yang, Y., Zhang, Z., & Hao, Z. (2016). Convex optimization for linear query processing under approximate differential privacy. In KDD 2016 - Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (Vol. 13-17-August-2016, pp. 2005-2014). Association for Computing Machinery. https://doi.org/10.1145/2939672.2939818

Convex optimization for linear query processing under approximate differential privacy. / Yuan, Ganzhao; Yang, Yin; Zhang, Zhenjie; Hao, Zhifeng.

KDD 2016 - Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Vol. 13-17-August-2016 Association for Computing Machinery, 2016. p. 2005-2014.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Yuan, G, Yang, Y, Zhang, Z & Hao, Z 2016, Convex optimization for linear query processing under approximate differential privacy. in KDD 2016 - Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. vol. 13-17-August-2016, Association for Computing Machinery, pp. 2005-2014, 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2016, San Francisco, United States, 13/8/16. https://doi.org/10.1145/2939672.2939818
Yuan G, Yang Y, Zhang Z, Hao Z. Convex optimization for linear query processing under approximate differential privacy. In KDD 2016 - Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Vol. 13-17-August-2016. Association for Computing Machinery. 2016. p. 2005-2014 https://doi.org/10.1145/2939672.2939818
Yuan, Ganzhao ; Yang, Yin ; Zhang, Zhenjie ; Hao, Zhifeng. / Convex optimization for linear query processing under approximate differential privacy. KDD 2016 - Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Vol. 13-17-August-2016 Association for Computing Machinery, 2016. pp. 2005-2014
@inproceedings{d0342230003a4d8d97a29821b114d843,
title = "Convex optimization for linear query processing under approximate differential privacy",
abstract = "Differential privacy enables organizations to collect accurate aggregates over sensitive data with strong, rigorous guarantees on individuals' privacy. Previous work has found that under differential privacy, computing multiple correlated aggregates as a batch, using an appropriate strategy, may yield higher accuracy than computing each of them independently. However, finding the best strategy that maximizes result accuracy is non-trivial, as it involves solving a complex constrained optimization program that appears to be non-convex. Hence, in the past much effort has been devoted in solving this non-convex optimization program. Existing approaches include various sophisticated heuristics and expensive numerical solutions. None of them, however, guarantees to find the optimal solution of this optimization problem. This paper points out that under (ϵ, δ)-differential privacy, the optimal solution of the above constrained optimization problem in search of a suitable strategy can be found, rather surprisingly, by solving a simple and elegant convex optimization program. Then, we propose an efficient algorithm based on Newton's method, which we prove to always converge to the optimal solution with linear global convergence rate and quadratic local convergence rate. Empirical evaluations demonstrate the accuracy and efficiency of the proposed solution.",
author = "Ganzhao Yuan and Yin Yang and Zhenjie Zhang and Zhifeng Hao",
year = "2016",
month = "8",
day = "13",
doi = "10.1145/2939672.2939818",
language = "English",
volume = "13-17-August-2016",
pages = "2005--2014",
booktitle = "KDD 2016 - Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining",
publisher = "Association for Computing Machinery",

}

TY - GEN

T1 - Convex optimization for linear query processing under approximate differential privacy

AU - Yuan, Ganzhao

AU - Yang, Yin

AU - Zhang, Zhenjie

AU - Hao, Zhifeng

PY - 2016/8/13

Y1 - 2016/8/13

N2 - Differential privacy enables organizations to collect accurate aggregates over sensitive data with strong, rigorous guarantees on individuals' privacy. Previous work has found that under differential privacy, computing multiple correlated aggregates as a batch, using an appropriate strategy, may yield higher accuracy than computing each of them independently. However, finding the best strategy that maximizes result accuracy is non-trivial, as it involves solving a complex constrained optimization program that appears to be non-convex. Hence, in the past much effort has been devoted in solving this non-convex optimization program. Existing approaches include various sophisticated heuristics and expensive numerical solutions. None of them, however, guarantees to find the optimal solution of this optimization problem. This paper points out that under (ϵ, δ)-differential privacy, the optimal solution of the above constrained optimization problem in search of a suitable strategy can be found, rather surprisingly, by solving a simple and elegant convex optimization program. Then, we propose an efficient algorithm based on Newton's method, which we prove to always converge to the optimal solution with linear global convergence rate and quadratic local convergence rate. Empirical evaluations demonstrate the accuracy and efficiency of the proposed solution.

AB - Differential privacy enables organizations to collect accurate aggregates over sensitive data with strong, rigorous guarantees on individuals' privacy. Previous work has found that under differential privacy, computing multiple correlated aggregates as a batch, using an appropriate strategy, may yield higher accuracy than computing each of them independently. However, finding the best strategy that maximizes result accuracy is non-trivial, as it involves solving a complex constrained optimization program that appears to be non-convex. Hence, in the past much effort has been devoted in solving this non-convex optimization program. Existing approaches include various sophisticated heuristics and expensive numerical solutions. None of them, however, guarantees to find the optimal solution of this optimization problem. This paper points out that under (ϵ, δ)-differential privacy, the optimal solution of the above constrained optimization problem in search of a suitable strategy can be found, rather surprisingly, by solving a simple and elegant convex optimization program. Then, we propose an efficient algorithm based on Newton's method, which we prove to always converge to the optimal solution with linear global convergence rate and quadratic local convergence rate. Empirical evaluations demonstrate the accuracy and efficiency of the proposed solution.

UR - http://www.scopus.com/inward/record.url?scp=84984996873&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84984996873&partnerID=8YFLogxK

U2 - 10.1145/2939672.2939818

DO - 10.1145/2939672.2939818

M3 - Conference contribution

VL - 13-17-August-2016

SP - 2005

EP - 2014

BT - KDD 2016 - Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

PB - Association for Computing Machinery

ER -