Memristor crossbar based unsupervised training
Memristor crossbar based unsupervised training
复制标题
基于忆阻器纵横杆的无监督训练
DOI：
10.1109/naecon.2015.7443091
复制
发表时间：
2015
期刊：
影响因子：
--
通讯作者：
T. Taha
中科院分区：
文献类型：
--
作者：
Raqibul Hasan;T. Taha
一些大数据应用程序特别关注分类和聚类任务。这种系统的鲁棒性取决于它从原始数据中提取重要特征的能力。对于大数据处理，我们感兴趣的是针对不同应用的通用特征提取机制。Autoencoder是一种流行的无监督训练算法，用于降维和特征提取。在这项工作中，我们研究了忆阻器交叉开关的实现自动编码器，这将消耗非常低的功率。我们设计了片上训练电路的无监督训练计划。
Several big data applications are particularly focused on classification and clustering tasks. Robustness of such system depends on how well it can extract important features from the raw data. For big data processing we are interested for a generic feature extraction mechanism for different applications. Autoencoder is a popular unsupervised training algorithm for dimensionality reduction and feature extraction. In this work we have examined memristor crossbar based implementation of autoencoder which will consume very low power. We have designed on-chip training circuitry for the unsupervised training scheme.