The Alpha-EM Algorithm: A Block Connectable Generalized Learning Tool for Neural Networks
The Alpha-EM Algorithm: A Block Connectable Generalized Learning Tool for Neural Networks
复制标题
Alpha-EM 算法：神经网络的块可连接广义学习工具
DOI：
10.1007/bfb0032507
复制
发表时间：
1997
期刊：
影响因子：
--
通讯作者：
Y. Matsuyama
中科院分区：
文献类型：
--
作者：
Y. Matsuyama
利用α-散度导出了一种广义期望最大化算法（EM算法）.该算法具有广泛的应用前景。本文主要研究混合概率的神经网络学习。α-EM算法包括现有的EM算法作为特殊情况，因为它对应于α= −1。参数α指定用于学习的概率权重。这个数字会影响学习速度和局部最优性。在讨论神经网络的更新方程时，还对Fisher有效得分、信息测度和Cramér-Rao不等式等基本统计量进行了推广。此外，本文还提出了另一个新的观点。研究发现，循环EM结构可以作为一个积木，以产生一个学习脉动阵列。将监视器连接到这个脉动阵列使得创建功能分布式学习系统成为可能。
Theα-divergence is utilized to derive a generalized expectation and maximization algorithm (EM algorithm). This algorithm has a wide range of applications. In this paper, neural network learning for mixture probabilities is focused. Theα-EM algorithm includes the existing EM algorithm as a special case since that corresponds toα= −1. The parameterαspecifies a probability weight for the learning. This number affects learning speed and local optimality. In the discussions of update equations of neural nets, extensions of basic statistics such as Fisher's efficient score, his measure of information and Cramér-Rao's inequality are also given. Besides, this paper unveils another new idea. It is found that the cyclic EM structure can be used as a building block to generate a learning systolic array. Attaching monitors to this systolic array makes it possible to create a functionally distributed learning system.
S.Amari：神经网络。
DOI：
--
发表时间：
--
期刊：
影响因子：
--
作者：
通讯作者：
--