计算机与数字工程2016，Vol.44Issue(8)：1436-1438,1547,4.DOI:10.3969/j.issn.1672-9722.2016.08.009
基于类别条件的受限玻尔兹曼机改进设计
Improvement of RBM Based on Label Condition
贺鹏程1
作者信息
- 1. 海军装备部驻重庆地区军事代表局 重庆 400042
- 折叠
摘要
Abstract
In order to overcome the poor generalization resulted by characteristic homogeneity for unsupervised training of restricted boltzmann machine ,the label condition is introduced to the training of RBM to design a new model named label condition RBM .Training for RBM ,the label information is treated as training condition for RBM and involved in the calcula-tion of hidden units’ posterior probability .The model is applied in deep learning which designed as the underlying structure of deep Boltzmann machine .Through a collection of handwritten numeral recognition test ,the model’s training speed and the effectiveness of feature extraction are greatly improved ,and the characterized learning ability of deep learning model is also improved .关键词
受限玻尔兹曼机/深度学习/监督学习/对比散度Key words
restricted Boltzmann machine/deep learning/supervised learning/contrastive divergence分类
信息技术与安全科学引用本文复制引用
贺鹏程..基于类别条件的受限玻尔兹曼机改进设计[J].计算机与数字工程,2016,44(8):1436-1438,1547,4.
