期刊
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS II-EXPRESS BRIEFS
卷 70, 期 6, 页码 1846-1850出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TCSII.2022.3231471
关键词
Memristors; Training; Computer architecture; Bayes methods; Deep learning; Multi-layer neural network; Task analysis; Memristor variations; ex situ training; Bayesian method; weight transfer; network robustness
Deep learning models implemented using memristors offer high scalability and energy efficiency for resource-constrained edge computing applications. However, the inherent physical randomness of memristors leads to significant performance degradation. In this study, a unified architecture incorporating a Bayesian-based training method and lightweight transfer scheme is proposed to address the robustness, energy, and time consumption issues caused by memristor variations. Experimental results demonstrate that this architecture can double the speed and energy efficiency of deploying deep learning models.
Deep learning models implemented using memristors show high scalability and high energy efficiency, promising a compact and efficient computing architecture for resource-constrained edge computing applications. These technologies integrate both data storage and computation simultaneously in a highly parallel memristor crossbar array architecture. However, the significant variations arising from the inherent physical randomness of memristors lead to a large performance degradation of deep learning models. The challenges of extensive energy costs and transfer time for deployment to maintain performance are faced. In this brief, for the first time, we propose a unified architecture that consists of a Bayesian-based training method and a lightweight transfer scheme. The proposed architecture can tackle the robustness, energy and time consumption issues caused by memristor variations. Our experimental results show that our architecture can double the speed and energy efficiency of deploying deep learning models.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据