paper-with-me

홈 › Papers

Entropy contraction of the Gibbs sampler under log-concavity

2024-10-01 · Filippo Ascolani, Hugo Lavenant, Giacomo Zanella

The Gibbs sampler (a.k.a. Glauber dynamics and heat-bath algorithm) is a popular Markov Chain Monte Carlo algorithm which iteratively samples from the conditional distributions of a probability measure $\pi$ of interest. Under the assumption that $\pi$ is strongly log-concave, we show that the random scan Gibbs sampler contracts in relative entropy and provide a sharp characterization of the associated contraction rate. Assuming that evaluating conditionals is cheap compared to evaluating the joint density, our results imply that the number of full evaluations of $\pi$ needed for the Gibbs sampler to mix grows linearly with the condition number and is independent of the dimension. If $\pi$ is non-strongly log-concave, the convergence rate in entropy degrades from exponential to polynomial. Our techniques are versatile and extend to Metropolis-within-Gibbs schemes and the Hit-and-Run algorithm. A comparison with gradient-based schemes and the connection with the optimization literature are also discussed.

📄 PDF Abstract BibTeX arXiv:2410.00858

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Spectral gap of Metropolis-within-Gibbs under log-concavity

2025-09-30 · Cecilia Secchi, Giacomo Zanella arxiv

The Metropolis-within-Gibbs (MwG) algorithm is a widely used Markov Chain Monte Carlo method for sampling from high-dimensional distributions when exact conditional sampling is intractable. We study MwG with Random Walk …

Rapid Mixing Swendsen-Wang Sampler for Stochastic Partitioned Attractive Models

2017-04-06 · Sejun Park, Yunhun Jang, Andreas Galanis, Jinwoo Shin 외

The Gibbs sampler is a particularly popular Markov chain used for learning and inference problems in Graphical Models (GMs). These tasks are computationally intractable in general, and the Gibbs sampler often suffers fro…

From Estimation to Sampling for Bayesian Linear Regression with Spike-and-Slab Prior

2023-07-09 · Qijia Jiang

We consider Bayesian linear regression with sparsity-inducing prior and design efficient sampling algorithms leveraging posterior contraction properties. A quasi-likelihood with Gaussian spike-and-slab (that is favorable…

regressionvalid

Improved analysis for a proximal algorithm for sampling

2022-02-13 · Yongxin Chen, Sinho Chewi, Adil Salim, Andre Wibisono

We study the proximal sampler of Lee, Shen, and Tian (2021) and obtain new convergence guarantees under weaker assumptions than strong log-concavity: namely, our results hold for (1) weakly log-concave targets, and (2) t…

Adaptive Scan Gibbs Sampler for Large Scale Inference Problems

2018-01-27 · Vadim Smolyakov, Qiang Liu, John W. Fisher III

For large scale on-line inference problems the update strategy is critical for performance. We derive an adaptive scan Gibbs sampler that optimizes the update frequency by selecting an optimum mini-batch size. We demonst…