摘要
本文讨论基于微分流形框架随机神经网络学习算法，称为ｅｍ学习算法；对于多层随机神经网络模型，我们从微分流形的角度分析它的对偶平坦流形结构，描述ｅｍ算法对于多层前馈随机神经网络模型学习算法实现和加速技术。ｅｍ算法可看作为在非完整数据处理下探索最大邻域估计与最好隐变量结构估计的对偶校正学习算法。
in this paper, a learning algorithm for stochastic neural networks, the em algorithm, is proposed based on differential manifold frameworks. At first, we analyze the dual flat manifold structure of stochastic neural networks, and then give the em algorithm implementation for 3 layers of stochastic perceptrons. A speeding theory of the em algorithm is shown. The em algorithm can be regarded as an iterative dual correction learning algorithm in the frameworks of imcomplete data estimation and the best model architecture choice.
出处
《计算机研究与发展》
EI
CSCD
北大核心
1996年第11期808-815,共8页
Journal of Computer Research and Development
关键词
随机神经网络
em学习算法
神经网络
Kullback-Leibler divergence, stochastic perceptron,em algorithm.
