The <i>em</i> algorithm for kernel matrix completion with auxiliary data

作     者：Tsuda, K Akaho, S Asai, K 

作者机构：Max Planck Inst Biol Cybernet D-72076 Tubingen Germany AIST Computat Biol Res Ctr Tokyo 1350064 Japan AIST Neurosci Res Inst Tsukuba Ibaraki 3058568 Japan Univ Tokyo Grad Sch Frontier Sci Dept Computat Biol Kashiwa Chiba 2778562 Japan 

出 版 物：《JOURNAL OF MACHINE LEARNING RESEARCH》 (机器学习研究杂志)

年 卷 期：2004年第4卷第1期

页      面：67-81页

核心收录：

学科分类：08[工学] 0811[工学-控制科学与工程] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

主　　题：information geometry em algorithm kernel matrix completion bacteria clustering 

摘      要：In biological data, it is often the case that observed data are available only for a subset of samples. When a kernel matrix is derived from such data, we have to leave the entries for unavailable samples as missing. In this paper, the missing entries are completed by exploiting an auxiliary kernel matrix derived from another information source. The parametric model of kernel matrices is created as a set of spectral variants of the auxiliary kernel matrix, and the missing entries are estimated by fitting this model to the existing entries. For model fitting, we adopt the em algorithm (distinguished from the EM algorithm of Dempster et al., 1977) based on the information geometry of positive definite matrices. We will report promising results on bacteria clustering experiments using two marker sequences: 16S and gyrB.