A new EM algorithm for flexibly tied GMMs with large number of components

Hadi Asheri,Reshad Hosseini,Babak Nadjar Araabi

Pattern Recognition（2021）

引用 19|浏览8

暂无评分

摘要

•A flexible tying scheme is used to solve the memory and computational loads of Gaussian mixture models.•To handle complexity of cost function, a fast Newton EM algorithm is proposed and is combined with a coordinate descent EM algorithm.•Computation factorization technique is proposed to increase the speed and decrease the memory requirements for the case of large number of components.

查看译文

关键词

Gaussian mixture model,Parameter sharing,Tied GMM,Computation factorization and reduction,Newton method,Fast minimal residual method,Clustering

AI 理解论文

溯源树

样例

生成溯源树，研究论文发展脉络

Chat Paper

正在生成论文摘要