Scholarlink

<p>We introduce a simple and efficient sampling strategy for the Dirichlet process mixture model (DPM) and its two-parameter extension, the Poisson-Dirichlet process mixture model, also known as the Pitman-Yor process mixture model (PYM). Inference in DPM and PYM is typically performed using Markov Chain Monte Carlo (MCMC) methods, specifically the Gibbs sampler. These sampling methods are usually divided into two classes: marginal and conditional algorithms. Each method has its own merits and limitations. The aim of this paper is to propose a simple and effective strategy that combines the main advantages of each class. Extensive experiments on simulated and real data highlight that the proposed sampler is relevant and performs much better than its competitors.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A simple and efficient method for sampling mixture models based on Dirichlet and Pitman-Yor processes

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年5月31日

查看全文
  • Mame Diarra Fall,
  • Éric Barat

摘要

We introduce a simple and efficient sampling strategy for the Dirichlet process mixture model (DPM) and its two-parameter extension, the Poisson-Dirichlet process mixture model, also known as the Pitman-Yor process mixture model (PYM). Inference in DPM and PYM is typically performed using Markov Chain Monte Carlo (MCMC) methods, specifically the Gibbs sampler. These sampling methods are usually divided into two classes: marginal and conditional algorithms. Each method has its own merits and limitations. The aim of this paper is to propose a simple and effective strategy that combines the main advantages of each class. Extensive experiments on simulated and real data highlight that the proposed sampler is relevant and performs much better than its competitors.