Convergence of Dirichlet Forms for MCMC Optimal Scaling with Dependent Target Distributions on Large Graphs
Abstract: Markov chain Monte Carlo (MCMC) algorithms have played a significant role in statistics, physics, machine learning and others, and they are the only known general and efficient approach for some high-dimensional problems. The random walk Metropolis (RWM) algorithm as the most classical MCMC algorithm, has had a great influence on the development and practice of science and engineering. The behavior of the RWM algorithm in high-dimensional problems is typically investigated through a weak convergence result of diffusion processes. In this paper, we utilize the Mosco convergence of Dirichlet forms in analyzing the RWM algorithm on large graphs, whose target distribution is the Gibbs measure that includes any probability measure satisfying a Markov property. The abstract and powerful theory of Dirichlet forms allows us to work directly and naturally on the infinite-dimensional space, and our notion of Mosco convergence allows Dirichlet forms associated with the RWM chains to lie on changing Hilbert spaces. Through the optimal scaling problem, we demonstrate the impressive strengths of the Dirichlet form approach over the standard diffusion approach.
- On partial integration in infinite-dimensional space and applications to Dirichlet forms. Journal of the London Mathematical Society, 2(1):122–136.
- An introduction to MCMC for machine learning. Machine learning, 50(1):5–43.
- On the geometric ergodicity of Metropolis-Hastings algorithms. Statistics, 41(1):77–84.
- Bédard, M. (2007). Weak convergence of Metropolis algorithms for non-i.i.d. target distributions. The Annals of Applied Probability, 17(4):1222–1244.
- Bédard, M. (2008). Optimal acceptance rates for Metropolis algorithms: Moving beyond 0.234. Stochastic Processes and their Applications, 118(12):2198–2222.
- Optimal scalings for local Metropolis–Hastings chains on nonproduct targets in high dimensions. The Annals of Applied Probability, 19(3):863–898.
- Billingsley, P. (2013). Convergence of probability measures. John Wiley & Sons.
- From Metropolis to diffusions: Gibbs states and optimal scaling. Stochastic Processes and their Applications, 90(2):181–206.
- Chen, M.-F. (2006). Eigenvalues, inequalities, and ergodic theory. Springer Science & Business Media.
- Doukhan, P. (2012). Mixing: properties and examples, volume 85. Springer Science & Business Media.
- Markov processes: characterization and convergence. John Wiley & Sons.
- Fritz, J. (1982). Stationary measures of stochastic gradient systems, infinite lattice models. Zeitschrift für Wahrscheinlichkeitstheorie und Verwandte Gebiete, 59(4):479–490.
- Weak convergence and optimal scaling of random walk Metropolis algorithms. The Annals of Applied Probability, 7(1):110–120.
- Georgii, H.-O. (2011). Gibbs measures and phase transitions. de Gruyter.
- Spectral gaps for a Metropolis–Hastings algorithm in infinite dimensions. The Annals of Applied Probability, 24(6):2455 – 2490.
- Kolesnikov, A. V. (2005). Convergence of Dirichlet forms with changing speed measures on ℝdsuperscriptℝ𝑑\mathbb{R}^{d}blackboard_R start_POSTSUPERSCRIPT italic_d end_POSTSUPERSCRIPT. Forum Mathematicum, 17(2):225–259.
- Convergence of spectral structures: a functional analytic theory and its applications to spectral geometry. Communications in analysis and geometry, 11(4):599–673.
- Introduction to the theory of (non-symmetric) Dirichlet forms. Springer Science & Business Media.
- Diffusion limits of the random walk Metropolis algorithm in high dimensions. The Annals of Applied Probability, 22(3):881–930.
- Stability of noisy Metropolis–Hastings. Statistics and Computing, 26(6):1187–1211.
- Mosco, U. (1994). Composite media and asymptotic Dirichlet forms. Journal of Functional Analysis, 123:368–421.
- Optimal scaling for partially updating MCMC algorithms. The Annals of Applied Probability, 16(2):475 – 515.
- Ergodic theorems for spatial processes. Zeitschrift für Wahrscheinlichkeitstheorie und verwandte Gebiete, 48(2):133–158.
- Optimal scaling of discrete approximations to Langevin diffusions. Journal of the Royal Statistical Society: Series B (Statistical Methodology), 60(1):255–268.
- Optimal scaling for various Metropolis-Hastings algorithms. Statistical science, 16(4):351–367.
- Optimal scaling of random-walk Metropolis algorithms on general target distributions. Stochastic Processes and their Applications, 130(10):6094–6132.
- A Dirichlet form approach to MCMC optimal scaling. Stochastic Processes and their Applications, 127(12):4053–4082.
Paper Prompts
Sign up for free to create and run prompts on this paper using GPT-5.
Top Community Prompts
Collections
Sign up for free to add this paper to one or more collections.