Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
120 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
46 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

Rosenthal-type inequalities for linear statistics of Markov chains (2303.05838v2)

Published 10 Mar 2023 in math.PR, math.ST, stat.ML, and stat.TH

Abstract: In this paper, we establish novel deviation bounds for additive functionals of geometrically ergodic Markov chains similar to Rosenthal and Bernstein inequalities for sums of independent random variables. We pay special attention to the dependence of our bounds on the mixing time of the corresponding chain. More precisely, we establish explicit bounds that are linked to the constants from the martingale version of the Rosenthal inequality, as well as the constants that characterize the mixing properties of the underlying Markov kernel. Finally, our proof technique is, up to our knowledge, new and based on a recurrent application of the Poisson decomposition.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (49)
  1. Radoslaw Adamczak. A tail inequality for suprema of unbounded empirical processes with applications to Markov chains. Electron. J. Probab., 13:no. 34, 1000–1034, 2008.
  2. Exponential concentration inequalities for additive functionals of Markov chains. ESAIM: Probability and Statistics, 19:440–481, 2015.
  3. A new approach to the limit theory of recurrent Markov chains. Transactions of the American Mathematical Society, 245:493–501, 1978.
  4. Concentration inequalities for sums and martingales. Springer, 2015.
  5. Bias and extrapolation in Markovian linear stochastic approximation with constant stepsizes. arXiv preprint arXiv:2210.00953, 2022.
  6. New Bernstein and Hoeffding type inequalities for regenerative Markov chains. Latin American journal of probability and mathematical statistics, 16:1–19, 02 2019.
  7. Stéphan JM Clémençon. Moment and probability inequalities for sums of bounded additive functionals of regular Markov chains via the Nummelin splitting technique. Statistics & probability letters, 55(3):227–238, 2001.
  8. Arnak S Dalalyan. Theoretical guarantees for approximate sampling from smooth and log-concave densities. Journal of the Royal Statistical Society. Series B (Statistical Methodology), pages 651–676, 2017.
  9. Subgaussian concentration inequalities for geometrically ergodic Markov chains. Electronic Communications in Probability, 20(none):1 – 12, 2015.
  10. Adapting to mixing time in stochastic optimization with Markovian data. In Kamalika Chaudhuri, Stefanie Jegelka, Le Song, Csaba Szepesvari, Gang Niu, and Sivan Sabato, editors, Proceedings of the 39th International Conference on Machine Learning, volume 162 of Proceedings of Machine Learning Research, pages 5429–5446. PMLR, 17–23 Jul 2022.
  11. Markov chains. Springer Series in Operations Research and Financial Engineering. Springer, 2018.
  12. A new weak dependence condition and applications to moment inequalities. Stochastic Process. Appl., 84(2):313–342, 1999.
  13. Probability and moment inequalities for sums of weakly dependent random variables, with applications. Stochastic Processes and their Applications, 117(7):878–903, 2007.
  14. Nonasymptotic convergence analysis for the unadjusted Langevin algorithm. Ann. Appl. Probab., 27(3):1551–1587, 06 2017.
  15. On the geometric convergence for mala under verifiable conditions. arXiv preprint arXiv:2201.01951, 2022.
  16. Probability and moment inequalities for additive functionals of geometrically ergodic Markov chains. arXiv preprint arXiv:2109.00331, 2021.
  17. Finite-time high-probability bounds for Polyak-Ruppert averaged iterates of linear stochastic approximation. arXiv preprint arXiv:2207.04475, 2022.
  18. Tight high probability bounds for linear stochastic approximation with fixed stepsize. In M. Ranzato, A. Beygelzimer, K. Nguyen, P. S. Liang, J. W. Vaughan, and Y. Dauphin, editors, Advances in Neural Information Processing Systems, volume 34, pages 30063–30074. Curran Associates, Inc., 2021.
  19. A. Eberle. Error bounds for Metropolis–Hastings algorithms applied to perturbations of Gaussian measures in high dimensions. The Annals of Applied Probability, 24(1):337–377, 02 2014.
  20. Hoeffding’s inequality for general Markov chains and its applications to statistical learning. J. Mach. Learn. Res., 22:139–1, 2021.
  21. Gersende Fort. Central limit theorems for stochastic approximation with controlled markov chain dynamics. ESAIM: Probability and Statistics, 19:60–80, 2015.
  22. Alison L. Gibbs. Convergence in the Wasserstein metric for Markov Chain Monte Carlo algorithms with applications to image restoration. Stochastic Models, 20(4):473–492, 2004.
  23. Hoeffding’s inequality for uniformly ergodic Markov chains. Statistics & probability letters, 56(2):143–146, 2002.
  24. Necessary and sufficient conditions for two classes of functions to be logarithmically completely monotonic. Integral Transforms and Special Functions, 18(11):819–826, 2007.
  25. M. Hairer. Ergodic theory for stochastic pdes. Unpublished lecture notes, 2008.
  26. Asymptotic coupling and a general form of Harris’ theorem with applications to stochastic delay equations. Probability Theory and Related Fields, 149(1-2):223–259, 2011.
  27. Spectral gaps for Metropolis-Hastings algorithms in infinite dimensions. Ann. Appl. Probab., 24:2455–290, 2014.
  28. A quantitative McDiarmid’s inequality for geometrically ergodic Markov chains. Electronic Communications in Probability, 25(none):1 – 11, 2020.
  29. Bernstein’s inequality for general Markov chains, 2018. arXiv pre-print arXiv:1805.10721.
  30. G. L. Jones. On the Markov chain central limit theorem. Probability Surveys, 1(none):299 – 320, 2004.
  31. A. Joulin and Y. Ollivier. Curvature, concentration and error estimates for Markov chain Monte Carlo. The Annals of Probability, 38(6):2418 – 2442, 2010.
  32. Relative entropy and exponential deviation bounds for general markov chains. In Proceedings. International Symposium on Information Theory, 2005. ISIT 2005., pages 1563–1567. IEEE, 2005.
  33. Geometric ergodicity and the spectral gap of non-reversible Markov chains. Probability Theory and Related Fields, 154(1-2):327–339, 2012.
  34. Michał Lemańczyk. General Bernstein-like inequality for additive functionals of Markov chains. Journal of Theoretical Probability, pages 1–29, 2020.
  35. On a method of semi-invariants. Theor. Probability Appl., 4:319–329, 1959.
  36. Probability inequalities. Springer Science & Business Media, 2011.
  37. S. Meyn and R. Tweedie. Markov Chains and Stochastic Stability. Cambridge University Press, New York, NY, USA, 2nd edition, 2009.
  38. Błażej Miasojedow. Hoeffding’s inequalities for geometrically ergodic markov chains on general state space. Statistics & Probability Letters, 87:115–120, 2014.
  39. On linear stochastic approximation: Fine-grained Polyak-Ruppert and non-asymptotic concentration. In Conference on Learning Theory, pages 2947–2997. PMLR, 2020.
  40. Optimal and instance-dependent guarantees for Markovian linear stochastic approximation. arXiv preprint arXiv:2112.12770, 2021.
  41. E. Nummelin. A splitting technique for harris recurrent Markov chains. Zeitschrift für Wahrscheinlichkeitstheorie und Verwandte Gebiete, 43:309–318, 1978.
  42. Daniel Paulin et al. Concentration inequalities for Markov chains by marton couplings and spectral methods. Electronic Journal of Probability, 20, 2015.
  43. Iosif Pinelis. Optimum Bounds for the Distributions of Martingales in Banach Spaces. The Annals of Probability, 22(4):1679 – 1706, 1994.
  44. Acceleration of stochastic approximation by averaging. SIAM journal on control and optimization, 30(4):838–855, 1992.
  45. General state space Markov chains and MCMC algorithms. Probab. Surv., 1:20–71, 2004.
  46. Exponential convergence of Langevin distributions and their discrete approximations. Bernoulli, 2(4):341–363, 1996.
  47. Haskell P. Rosenthal. On the subspaces of Lpsuperscript𝐿𝑝L^{p}italic_L start_POSTSUPERSCRIPT italic_p end_POSTSUPERSCRIPT (p>2)𝑝2(p>2)( italic_p > 2 ) spanned by sequences of independent random variables. Israel J. Math., 8:273–303, 1970.
  48. David Ruppert. Efficient estimations from a slowly convergent robbins-monro process. Technical report, Cornell University Operations Research and Industrial Engineering, 1988.
  49. L. Saulis and V. A. Statulevičius. Limit theorems for large deviations, volume 73 of Mathematics and its Applications (Soviet Series). Kluwer Academic Publishers Group, Dordrecht, 1991. Translated and revised from the 1989 Russian original.
Citations (3)
View on Semantic Scholar

Summary

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

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