Private Learning and Sanitization: Pure vs. Approximate Differential Privacy

  • Amos Beimel
  • Kobbi Nissim
  • Uri Stemmer
Part of the Lecture Notes in Computer Science book series (LNCS, volume 8096)

Abstract

We compare the sample complexity of private learning and sanitization tasks under pureε-differential privacy [Dwork, McSherry, Nissim, and Smith TCC 2006] and approximate (ε,δ)-differential privacy [Dwork, Kenthapadi, McSherry, Mironov, and Naor EUROCRYPT 2006]. We show that the sample complexity of these tasks under approximate differential privacy can be significantly lower than that under pure differential privacy.

Keywords

Differential Privacy Private Learning Sanitization 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Beimel, A., Kasiviswanathan, S.P., Nissim, K.: Bounds on the sample complexity for private learning and private data release. In: Micciancio, D. (ed.) TCC 2010. LNCS, vol. 5978, pp. 437–454. Springer, Heidelberg (2010)CrossRefGoogle Scholar
  2. 2.
    Beimel, A., Nissim, K., Stemmer, U.: Characterizing the sample complexity of private learners. In: ITCS, pp. 97–110 (2013)Google Scholar
  3. 3.
    Blum, A., Dwork, C., McSherry, F., Nissim, K.: Practical privacy: The SuLQ framework. In: PODS, pp. 128–138. ACM (2005)Google Scholar
  4. 4.
    Blum, A., Ligett, K., Roth, A.: A learning theory approach to noninteractive database privacy. J. ACM 60(2), 12:1–12:25 (2013)Google Scholar
  5. 5.
    Chaudhuri, K., Hsu, D.: Sample complexity bounds for differentially private learning. In: COLT, vol. 19, pp. 155–186 (2011)Google Scholar
  6. 6.
    De, A.: Lower bounds in differential privacy. In: Cramer, R. (ed.) TCC 2012. LNCS, vol. 7194, pp. 321–338. Springer, Heidelberg (2012)CrossRefGoogle Scholar
  7. 7.
    Dwork, C., Kenthapadi, K., McSherry, F., Mironov, I., Naor, M.: Our data, ourselves: Privacy via distributed noise generation. In: Vaudenay, S. (ed.) EUROCRYPT 2006. LNCS, vol. 4004, pp. 486–503. Springer, Heidelberg (2006)CrossRefGoogle Scholar
  8. 8.
    Dwork, C., Lei, J.: Differential privacy and robust statistics. In: STOC 2009, pp. 371–380. ACM, New York (2009)Google Scholar
  9. 9.
    Dwork, C., McSherry, F., Nissim, K., Smith, A.: Calibrating noise to sensitivity in private data analysis. In: Halevi, S., Rabin, T. (eds.) TCC 2006. LNCS, vol. 3876, pp. 265–284. Springer, Heidelberg (2006)CrossRefGoogle Scholar
  10. 10.
    Dwork, C., Naor, M., Reingold, O., Rothblum, G., Vadhan, S.: On the complexity of differentially private data release. In: STOC, pp. 381–390. ACM (2009)Google Scholar
  11. 11.
    Gupta, A., Hardt, M., Roth, A., Ullman, J.: Privately releasing conjunctions and the statistical query barrier. In: STOC, pp. 803–812. ACM, New York (2011)Google Scholar
  12. 12.
    Hardt, M., Talwar, K.: On the geometry of differential privacy. In: STOC, pp. 705–714 (2010)Google Scholar
  13. 13.
    Hardt, M.A.W.: A Study of Privacy and Fairness in Sensitive Data Analysis. PhD thesis, Princeton University (2011)Google Scholar
  14. 14.
    Kasiviswanathan, S.P., Lee, H.K., Nissim, K., Raskhodnikova, S., Smith, A.: What can we learn privately? In: FOCS, pp. 531–540. IEEE Computer Society (2008)Google Scholar
  15. 15.
    Kearns, M.J.: Efficient noise-tolerant learning from statistical queries. Journal of the ACM 45(6), 983–1006 (1998); Preliminary version in Proceedings of STOC 1993Google Scholar
  16. 16.
    McSherry, F., Talwar, K.: Mechanism design via differential privacy. In: FOCS, pp. 94–103. IEEE (2007)Google Scholar
  17. 17.
    Roth, A.: Differential privacy and the fat-shattering dimension of linear queries. In: Serna, M., Shaltiel, R., Jansen, K., Rolim, J. (eds.) APPROX and RANDOM 2010. LNCS, vol. 6302, pp. 683–695. Springer, Heidelberg (2010)CrossRefGoogle Scholar
  18. 18.
    Smith, A., Thakurta, A.: Differentially private feature selection via stability arguments, and the robustness of the lasso. Manuscript (2012)Google Scholar
  19. 19.
    Ullman, J.: Answering n2 + o(1) counting queries with differential privacy is hard. CoRR, abs/1207.6945 (2012)Google Scholar
  20. 20.
    Ullman, J., Vadhan, S.: PCPs and the hardness of generating private synthetic data. In: Ishai, Y. (ed.) TCC 2011. LNCS, vol. 6597, pp. 400–416. Springer, Heidelberg (2011)CrossRefGoogle Scholar
  21. 21.
    Valiant, L.G.: A theory of the learnable. Communications of the ACM 27, 1134–1142 (1984)MATHCrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2013

Authors and Affiliations

  • Amos Beimel
    • 1
  • Kobbi Nissim
    • 1
    • 2
  • Uri Stemmer
    • 1
  1. 1.Dept. of Computer ScienceBen-Gurion UniversityIsrael
  2. 2.Dept. of Computer ScienceHarvard UniversityIsrael

Personalised recommendations