[go: up one dir, main page]
More Web Proxy on the site http://driver.im/ skip to main content
10.1145/2808769.2808775acmconferencesArticle/Chapter ViewAbstractPublication PagesccsConference Proceedingsconference-collections
research-article

Differential Privacy for Classifier Evaluation

Published: 16 October 2015 Publication History

Abstract

Differential privacy provides powerful guarantees that individuals incur minimal additional risk by including their personal data in a database. Most work in differential privacy has focused on differentially private algorithms that produce models, counts, and histograms. Nevertheless, even with a classification model produced by a differentially private algorithm, directly reporting the classifier's performance on a database has the potential for disclosure. Thus, differentially private computation of evaluation metrics for machine learning is an important research area. We find effective mechanisms for area under the receiver-operating characteristic (ROC) curve and average precision.

References

[1]
K. Bache and M. Lichman. UCI machine learning repository, 2013.
[2]
K. Boyd, V. S. Costa, J. Davis, and D. Page. Unachievable region in precision-recall space and its effect on empirical evaluation. In ICML, pages 639--646, 2012.
[3]
K. Chaudhuri, C. Monteleoni, and A. D. Sarwate. Differentially private empirical risk minimization. The Journal of Machine Learning Research, 12:1069--1109, 2011.
[4]
K. Chaudhuri and S. A. Vinterbo. A stability-based validation procedure for differentially private machine learning. In NIPS, pages 2652--2660, 2013.
[5]
C. Dwork. Differential privacy. In ICALP. Springer, 2006.
[6]
C. Dwork, F. McSherry, K. Nissim, and A. Smith. Calibrating noise to sensitivity in private data analysis. In Theory of Cryptography, pages 265--284. Springer, 2006.
[7]
A. Friedman and A. Schuster. Data mining with differential privacy. In KDD, pages 493--502. ACM, 2010.
[8]
A. Ghosh, T. Roughgarden, and M. Sundararajan. Universally utility-maximizing privacy mechanisms. In STOC, 2009.
[9]
R. Hall, A. Rinaldo, and L. Wasserman. Differential privacy for functions and functional data. The Journal of Machine Learning Research, 14(1):703--727, 2013.
[10]
D. Kifer and A. Machanavajjhala. Pufferfish: A framework for mathematical privacy definitions. ACM Trans. Database Syst., 39(1):3:1--3:36, Jan. 2014.
[11]
C. D. Manning, P. Raghavan, and H. Schütze. Introduction to Information Retrieval. Cambridge University Press, New York, NY, USA, 2008.
[12]
G. J. Matthews and O. Harel. An examination of data confidentiality and disclosure issues related to publication of empiricalROC\ curves. Academic Radiology, 20(7):889 -- 896, 2013.
[13]
K. Nissim, S. Raskhodnikova, and A. Smith. Smooth sensitivity and sampling in private data analysis. In STOC, page 75. ACM Press, 2007.
[14]
M. S. Pepe. The statistical evaluation of medical tests for classification and prediction. Oxford University Press, USA, 2004.
[15]
F. J. Provost, T. Fawcett, and R. Kohavi. The case against accuracy estimation for comparing induction algorithms. In ICML, volume 98, pages 445--453, 1998.
[16]
B. I. Rubinstein, P. L. Bartlett, L. Huang, and N. Taft. Learning in a large function space: Privacy-preserving mechanisms for svm learning. preprint arXiv:0911.5708, 2009.
[17]
B. Stoddard, Y. Chen, and A. Machanavajjhala. Differentially private algorithms for empirical machine learning. preprint arXiv:1411.5428, 2014.
[18]
L. Wasserman and S. Zhou. A statistical framework for differential privacy. Journal of the American Statistical Association, 105(489):375--389, 2010.
[19]
J. Zhang, Z. Zhang, X. Xiao, Y. Yang, and M. Winslett. Functional mechanism: regression analysis under differential privacy. VLDB, 5(11):1364--1375, 2012.

Cited By

View all
  • (2024)Distributed non-disclosive validation of predictive models by a modified ROC-GLMBMC Medical Research Methodology10.1186/s12874-024-02312-424:1Online publication date: 29-Aug-2024
  • (2024)Benchmarking Evaluation Protocols for Classifiers Trained on Differentially Private Synthetic DataIEEE Access10.1109/ACCESS.2024.344691312(118637-118648)Online publication date: 2024
  • (2024)An Efficient Data Privacy Protection System Based on Differential PrivacyData Science and Communication10.1007/978-981-99-5435-3_58(785-798)Online publication date: 3-Jan-2024
  • Show More Cited By

Recommendations

Comments

Please enable JavaScript to view thecomments powered by Disqus.

Information & Contributors

Information

Published In

cover image ACM Conferences
AISec '15: Proceedings of the 8th ACM Workshop on Artificial Intelligence and Security
October 2015
110 pages
ISBN:9781450338264
DOI:10.1145/2808769
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than the author(s) must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected].

Sponsors

Publisher

Association for Computing Machinery

New York, NY, United States

Publication History

Published: 16 October 2015

Permissions

Request permissions for this article.

Check for updates

Author Tags

  1. ROC curve
  2. average precision
  3. differential privacy

Qualifiers

  • Research-article

Funding Sources

  • US National Library of Medicine

Conference

CCS'15
Sponsor:

Acceptance Rates

AISec '15 Paper Acceptance Rate 11 of 25 submissions, 44%;
Overall Acceptance Rate 94 of 231 submissions, 41%

Upcoming Conference

CCS '25

Contributors

Other Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

  • Downloads (Last 12 months)37
  • Downloads (Last 6 weeks)2
Reflects downloads up to 22 Dec 2024

Other Metrics

Citations

Cited By

View all
  • (2024)Distributed non-disclosive validation of predictive models by a modified ROC-GLMBMC Medical Research Methodology10.1186/s12874-024-02312-424:1Online publication date: 29-Aug-2024
  • (2024)Benchmarking Evaluation Protocols for Classifiers Trained on Differentially Private Synthetic DataIEEE Access10.1109/ACCESS.2024.344691312(118637-118648)Online publication date: 2024
  • (2024)An Efficient Data Privacy Protection System Based on Differential PrivacyData Science and Communication10.1007/978-981-99-5435-3_58(785-798)Online publication date: 3-Jan-2024
  • (2023)A Differential Privacy-based System for Efficiently Protecting Data Privacy2023 International Conference on Sustainable Computing and Smart Systems (ICSCSS)10.1109/ICSCSS57650.2023.10169412(1399-1404)Online publication date: 14-Jun-2023
  • (2023)AURA: Privacy-Preserving Augmentation to Improve Test Set Diversity in Speech EnhancementICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)10.1109/ICASSP49357.2023.10096879(1-5)Online publication date: 4-Jun-2023
  • (2023)Privacy-preserving and lossless distributed estimation of high-dimensional generalized additive mixed modelsStatistics and Computing10.1007/s11222-023-10323-234:1Online publication date: 7-Nov-2023
  • (2023)ppAURORA: Privacy Preserving Area Under Receiver Operating Characteristic and Precision-Recall CurvesNetwork and System Security10.1007/978-3-031-39828-5_15(265-280)Online publication date: 7-Aug-2023
  • (2021)Reducing bias and increasing utility by federated generative modeling of medical images using a centralized adversaryProceedings of the Conference on Information Technology for Social Good10.1145/3462203.3475875(79-84)Online publication date: 9-Sep-2021
  • (2019)Privacy Accounting and Quality Control in the Sage Differentially Private ML PlatformACM SIGOPS Operating Systems Review10.1145/3352020.335203253:1(75-84)Online publication date: 25-Jul-2019
  • (2019)Privacy accounting and quality control in the sage differentially private ML platformProceedings of the 27th ACM Symposium on Operating Systems Principles10.1145/3341301.3359639(181-195)Online publication date: 27-Oct-2019
  • Show More Cited By

View Options

Login options

View options

PDF

View or Download as a PDF file.

PDF

eReader

View online with eReader.

eReader

Media

Figures

Other

Tables

Share

Share

Share this Publication link

Share on social media