Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
156 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
45 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Differentially Private High Dimensional Bandits (2402.03737v1)

Published 6 Feb 2024 in cs.LG, cs.CR, cs.SY, eess.SY, math.OC, and stat.ML

Abstract: We consider a high-dimensional stochastic contextual linear bandit problem when the parameter vector is $s_{0}$-sparse and the decision maker is subject to privacy constraints under both central and local models of differential privacy. We present PrivateLASSO, a differentially private LASSO bandit algorithm. PrivateLASSO is based on two sub-routines: (i) a sparse hard-thresholding-based privacy mechanism and (ii) an episodic thresholding rule for identifying the support of the parameter $\theta$. We prove minimax private lower bounds and establish privacy and utility guarantees for PrivateLASSO for the central model under standard assumptions.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (49)
  1. Improved algorithms for linear stochastic bandits. Advances in neural information processing systems, 24.
  2. Agrawal, R. (1995). Sample mean based index policies by o (log n) regret for the multi-armed bandit problem. Advances in Applied Probability, 27(4):1054–1078.
  3. Thompson sampling for contextual bandits with linear payoffs. In International conference on machine learning, pages 127–135. PMLR.
  4. Thresholded lasso bandit. arXiv preprint arXiv:2010.11994.
  5. Fast learning rates for plug-in classifiers. The Annals of statistics, 35(2):608–633.
  6. Finite-time analysis of the multiarmed bandit problem. Machine learning, 47(2):235–256.
  7. Local, private, efficient protocols for succinct histograms. In Proceedings of the forty-seventh annual ACM symposium on Theory of computing, pages 127–135.
  8. Online decision making with high-dimensional covariates. Operations Research, 68(1):276–294.
  9. Least squares after model selection in high-dimensional sparse models.
  10. Statistics for high-dimensional data: methods, theory and applications. Springer Science & Business Media.
  11. The cost of privacy in generalized linear models: Algorithms and minimax lower bounds. arXiv preprint arXiv:2011.03900.
  12. Bandit theory meets compressed sensing for high dimensional stochastic linear bandit. In Artificial Intelligence and Statistics, pages 190–198. PMLR.
  13. Private and continual release of statistics. ACM Transactions on Information and System Security (TISSEC), 14(3):1–24.
  14. Differentially private empirical risk minimization. Journal of Machine Learning Research, 12(3).
  15. Privacy-preserving dynamic personalized pricing with demand learning. Management Science.
  16. The possibilities and limitations of private prediction markets. ACM Transactions on Economics and Computation, 8(3).
  17. Gamification of pure exploration for linear bandits. In International Conference on Machine Learning, pages 2432–2442. PMLR.
  18. Calibrating noise to sensitivity in private data analysis. In Proceedings of the 3rd Conference on Theory of Cryptography, TCC ’06, pages 265–284.
  19. Differential privacy under continual observation. In Proceedings of the forty-second ACM symposium on Theory of computing, pages 715–724.
  20. On the complexity of differentially private data release: efficient algorithms and hardness results. In Proceedings of the 41st ACM Symposium on Theory of Computing, STOC ’09, pages 381–390.
  21. The algorithmic foundations of differential privacy. Foundations and Trends in Theoretical Computer Science, 9(3–4):211–407.
  22. Boosting and differential privacy. In Foundations of Computer Science (FOCS), 2010 51st Annual IEEE Symposium on. IEEE, pages 51–60.
  23. Analyze gauss: optimal bounds for privacy-preserving principal component analysis. In Proceedings of the forty-sixth annual ACM symposium on Theory of computing, pages 11–20.
  24. A linear response bandit problem. Stochastic Systems, 3(1):230–261.
  25. A general framework to analyze stochastic linear bandit. arXiv preprint arXiv:2002.05152.
  26. Generalized linear bandits with local differential privacy. Advances in Neural Information Processing Systems, 34.
  27. Differentially private stochastic linear bandits:(almost) for free. arXiv preprint arXiv:2207.03445.
  28. Adaptive exploration in linear contextual bandit. In International Conference on Artificial Intelligence and Statistics, pages 3536–3545. PMLR.
  29. High-dimensional sparse linear bandits. Advances in Neural Information Processing Systems, 33:10753–10763.
  30. Private matchings and allocations. In Proceedings of the Forty-sixth Annual ACM Symposium on Theory of Computing, STOC ’14, pages 21–30.
  31. Mechanism design in large games: Incentives and privacy. In Proceedings of the 5th Conference on Innovations in Theoretical Computer Science, ITCS ’14, pages 403–410. ACM.
  32. Private convex empirical risk minimization and high-dimensional regression. In Conference on Learning Theory, pages 25–1. JMLR Workshop and Conference Proceedings.
  33. Asymptotically efficient adaptive allocation rules. Advances in applied mathematics, 6(1):4–22.
  34. Bandit algorithms. Cambridge University Press.
  35. A simple unified framework for high dimensional bandit problems. arXiv preprint arXiv:2102.09626.
  36. (nearly) optimal differentially private stochastic multi-arm bandits. In Proceedings of the Thirty-First Conference on Uncertainty in Artificial Intelligence, pages 592–601.
  37. Sparsity-agnostic lasso bandit. In International Conference on Machine Learning, pages 8271–8280. PMLR.
  38. Multi-armed bandits with local differential privacy. arXiv preprint arXiv:2007.03121.
  39. Asymptotically truthful equilibrium selection in large congestion games. In Proceedings of the Fifteenth ACM Conference on Economics and Computation, EC ’14, pages 771–782.
  40. Differentially private contextual linear bandits. Advances in Neural Information Processing Systems, 31.
  41. Sheffet, O. (2015). Private approximations of the 2nd-moment matrix using existing techniques in linear regression. arXiv preprint arXiv:1507.00056.
  42. Nearly optimal private lasso. Advances in Neural Information Processing Systems, 28.
  43. (nearly) optimal algorithms for private online learning in full-information and bandit settings. In Burges, C. J. C., Bottou, L., Welling, M., Ghahramani, Z., and Weinberger, K. Q., editors, Advances in Neural Information Processing Systems, volume 26. Curran Associates, Inc.
  44. Tibshirani, R. (1996). Regression shrinkage and selection via the lasso. Journal of the Royal Statistical Society: Series B (Methodological), 58(1):267–288.
  45. Tsybakov, A. B. (2004). Optimal aggregation of classifiers in statistical learning. The Annals of Statistics, 32(1):135–166.
  46. High dimensional sparse linear regression under local differential privacy: Power and limitations. In 2018 NIPS workshop in Privacy-Preserving Machine Learning, volume 235.
  47. On sparse linear regression in the local differential privacy model. In International Conference on Machine Learning, pages 6628–6637. PMLR.
  48. Paprika: Private online false discovery rate control. arXiv preprint arXiv:2002.12321.
  49. Locally differentially private (contextual) bandits learning. Advances in Neural Information Processing Systems, 33:12300–12310.

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now