Computational efficient Variational Bayesian Gaussian Mixture Models via Coreset
Computational efficient Variational Bayesian Gaussian Mixture Models via Coreset
复制标题
通过 Coreset 计算高效的变分贝叶斯高斯混合模型
DOI：
--
复制
发表时间：
2016
期刊：
影响因子：
--
通讯作者：
Teresa Wu
中科院分区：
文献类型：
--
作者：
Min Zhang;Yinlin Fu;K. Bennett;Teresa Wu
变分贝叶斯-高斯混合模型是目前流行的一种性能可靠的聚类算法。然而，需要注意的是，模型拟合过程需要很长时间，特别是在处理大规模数据时，因为它使用了整个数据集。为了解决这一问题，本文提出了一种基于Coreset的加权VBGMM算法。具体而言，首先提出了一种新的核心集构建方法来对用于拟合模型的数据进行采样。为了评估该算法，使用了两个数据集：1)六个大鼠肾脏图像数据集；2)三个人肾脏图像数据集。结果表明，本文提出的算法比经典的VBGMM要快得多（约20倍），同时在整个数据集上保持相似的性能。
Variational Bayesian Gaussian Mixture Model is a popular clustering algorithm with a reliable performance. However, it is noted that the model fitting process takes long time, especially when dealing with large scale data, since it utilizes the whole dataset. To address this issue, in paper we propose a new algorithm termed a weighted VBGMM via Coreset. Specifically, a new coreset construction method is first proposed to sample the data which is used to fit the model. To evaluate the algorithm, two datasets are used: 1) six rat kidney images datasets 2) three human kidney images datasets. The results show that our proposed algorithm is much faster (~ 20 times) comparing to classic VBGMM while maintaining the similar performance on whole dataset.