Space alternating data augmentation: application to finite mixture of Gaussians and speaker recognition
Space alternating data augmentation: application to finite mixture of Gaussians and speaker recognition
复制标题
空间交替数据增强：在高斯有限混合和说话人识别中的应用
DOI：
10.1109/icassp.2005.1416108
复制
发表时间：
2005
期刊：
影响因子：
--
通讯作者：
T. Matsui
中科院分区：
文献类型：
--
作者：
A. Doucet;S. Sénécal;T. Matsui
SAGE（空间交替广义期望最大化）算法（Celeux，G.例如，2001）是EM（期望最大化）算法的最优雅和最流行的扩展之一，用于执行ML（最大似然）或MAP（最大后验）参数估计。该算法通过交替丢失数据空间，以子块为单位更新参数分量。其效率已在许多模拟研究中得到报道。我们在这里提出了一个MCMC（马尔可夫链蒙特卡罗）策略命名为SADA（空间交替数据增强），它依赖于相同的原则，以有效地从（后验）分布采样，我们讨论了它的应用有限混合高斯。对于这个模型，我们还提出了一个原始的SAGE算法的实现。在蒙特卡洛模拟和说话人识别的应用中，这些方法是对标准EM和DA（数据增强）算法的直接修改，始终优于它们。
The SAGE (space-alternating generalized expectation-maximization) algorithm (Celeux, G. et al., 2001) is one of the most elegant and popular extensions of the EM (expectation maximization) algorithm for performing ML (maximum likelihood) or MAP (maximum a posteriori) parameter estimation. This algorithm updates parameter components by subblocks by alternating missing data spaces. Its efficiency has been reported in numerous simulation studies. We propose here an MCMC (Markov chain Monte Carlo) strategy named SADA (space-alternating data augmentation) which relies on the same principle in order to sample efficiently from (posterior) distributions and we discuss its application to finite mixtures of Gaussians. For this model, we also present an original implementation of the SAGE algorithm. In Monte Carlo simulations and in an application for speaker recognition, these methods, which are straightforward modifications of the standard EM and DA (data augmentation) algorithms, consistently outperform them.