Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
133 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

Spectral gap bounds for reversible hybrid Gibbs chains (2312.12782v4)

Published 20 Dec 2023 in math.ST, math.PR, and stat.TH

Abstract: Hybrid Gibbs samplers represent a prominent class of approximated Gibbs algorithms that utilize Markov chains to approximate conditional distributions, with the Metropolis-within-Gibbs algorithm standing out as a well-known example. Despite their widespread use in both statistical and non-statistical applications, little is known about their convergence properties. This article introduces novel methods for establishing bounds on the convergence rates of certain reversible hybrid Gibbs samplers. In particular, we examine the convergence characteristics of hybrid random-scan Gibbs algorithms. Our analysis reveals that the absolute spectral gap of a hybrid Gibbs chain can be bounded based on the absolute spectral gap of the exact Gibbs chain and the absolute spectral gaps of the Markov chains employed for conditional distribution approximations. We also provide a convergence bound of similar flavors for hybrid data augmentation algorithms, extending existing works on the topic. The general bounds are applied to three examples: a random-scan Metropolis-within-Gibbs sampler, random-scan Gibbs samplers with block updates, and a hybrid slice sampler.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (47)
  1. Spectral independence in high-dimensional expanders and applications to the hardcore model. SIAM Journal on Computing FOCS20–1.
  2. Comparison of Markov chains via weak Poincaré inequalities with application to pseudo-marginal MCMC. Annals of Statistics 50 3592–3618.
  3. Explicit convergence bounds for Metropolis Markov chains: isoperimetry, spectral gaps and profiles. arXiv preprint.
  4. A Short Course on Spectral Theory, vol. 209. Springer.
  5. Atchadé, Y. F. (2021). Approximate spectral gaps for Markov chain mixing times in high dimensions. SIAM Journal on Mathematics of Data Science 3 854–872.
  6. Baxendale, P. H. (2005). Renewal theory and computable convergence rates for geometrically ergodic Markov chains. Annals of Applied Probability 15 700–738.
  7. On the computational complexity of MCMC-based estimators in large samples. Annals of Statistics 2011–2055.
  8. On mixing of Markov chains: Coupling, spectral independence, and entropy factorization. In Proceedings of the 2022 Annual ACM-SIAM Symposium on Discrete Algorithms (SODA). SIAM.
  9. Determination of the spectral gap for Kac’s master equation and related stochastic evolution. Acta Mathematica 191 1–54.
  10. Explaining the Gibbs sampler. The American Statistician 46 167–174.
  11. Conditioning as disintegration. Statistica Neerlandica 51 287–317.
  12. Improved analysis for a proximal algorithm for sampling. In Conference on Learning Theory. PMLR.
  13. Fast mixing of Metropolized Hamiltonian Monte Carlo: Benefits of multi-step gradients. The Journal of Machine Learning Research 21 3647–3717.
  14. Optimal mixing of Glauber dynamics: Entropy factorization via high-dimensional expansion. In Proceedings of the 53rd Annual ACM SIGACT Symposium on Theory of Computing.
  15. Understanding the Metropolis-Hastings algorithm. The American Statistician 49 327–335.
  16. Log-concave sampling: Metropolis-hastings algorithms are fast. Journal of Machine Learning Research 20 1–42.
  17. Rapid mixing from spectral independence beyond the Boolean domain. ACM Transactions on Algorithms (TALG) 18 1–32.
  18. On the geometric ergodicity of hybrid samplers. Journal of Applied Probability 40 123–146.
  19. Stochastic relaxation, Gibbs distributions, and the Bayesian restoration of images. IEEE Transactions on pattern analysis and machine intelligence 721–741.
  20. Adaptive rejection Metropolis sampling within Gibbs sampling. Journal of the Royal Statistical Society Series C: Applied Statistics 44 455–472.
  21. Hobert, J. P. (2011). The data augmentation algorithm: Theory and methodology. In Handbook of Markov Chain Monte Carlo (S. Brooks, A. Gelman, G. Jones and X.-L. Meng, eds.). Chapman & Hall/CRC Press.
  22. Convergence of conditional Metropolis-Hastings samplers. Advances in Applied Probability 46 422–445.
  23. Data augmentation MCMC for Bayesian inference from privatized data. Advances in neural information processing systems 35 12732–12743.
  24. Convergence of hybrid slice sampling via spectral gap. arXiv preprint.
  25. Bounds on the l2superscript𝑙2l^{2}italic_l start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT spectrum for Markov chains and Markov processes: A generalization of Cheeger’s inequality. Transactions of the American Mathematical Society 309 557–580.
  26. Structured logconcave sampling with a restricted Gaussian oracle. In Conference on Learning Theory. PMLR.
  27. Structured logconcave sampling with a restricted Gaussian oracle. arXiv preprint.
  28. Liu, J. S. (1996). Peskun’s theorem and a modified discrete-state Gibbs sampler. Biometrika 83.
  29. Random walks in a convex body and an improved volume algorithm. Random structures & algorithms 4 359–412.
  30. Markov chain decomposition for convergence rate analysis. Annals of Applied Probability 581–606.
  31. Leveraging the exact likelihood of deep latent variable models. Advances in Neural Information Processing Systems 31.
  32. Equation of state calculations by fast computing machines. The journal of chemical physics 21 1087–1092.
  33. Quantitative spectral gap estimate and Wasserstein contraction of simple slice sampling. Annals of Applied Probability 31 806–825.
  34. Variable-at-a-time implementations of Metropolis-Hastings. arXiv preprint.
  35. Ollivier, Y. (2009). Ricci curvature of Markov chains on metric spaces. Journal of Functional Analysis 256 810–864.
  36. Pollard, D. (2002). A User’s Guide to Measure Theoretic Probability. 8, Cambridge University Press.
  37. Qin, Q. (2023). Geometric ergodicity of trans-dimensional Markov chain Monte Carlo algorithms. arXiv preprint.
  38. Convergence rates of two-component MCMC samplers. Bernoulli 28 859–885.
  39. Spectral telescope: Convergence rate bounds for random-scan Gibbs samplers based on a hierarchical structure. arXiv preprint.
  40. Geometric ergodicity and hybrid Markov chains. Electronic Communications in Probability 2 13–25.
  41. Two convergence properties of hybrid samplers. Annals of Applied Probability 8 397–407.
  42. Geometric L2superscript𝐿2L^{2}italic_L start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT and L1superscript𝐿1L^{1}italic_L start_POSTSUPERSCRIPT 1 end_POSTSUPERSCRIPT convergence are equivalent for reversible Markov chains. Journal of Applied Probability 38 37–41.
  43. Bayesian inference for latent factor copulas and application to financial risk forecasting. Econometrics 5.
  44. The calculation of posterior distributions by data augmentation (with discussion). Journal of the American Statistical Association 82 528–540.
  45. Tierney, L. (1998). A note on Metropolis-Hastings kernels for general state spaces. Annals of Applied Probability 1–9.
  46. The art of data augmentation (with discussion). Journal of Computational and Graphical Statistics 10 1–50.
  47. Bayesian non-homogeneous hidden Markov model with variable selection for investigating drivers of seizure risk cycling. Annals of Applied Statistics 17 333 – 356.
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