基于DHNN网络的玻尔兹曼机权值计算研究  

Research onWeight Calculation of Boltzmann Machine Based on Hopfield Network

作　　者：张波[1] ZHANG Bo(Changzhou Vocational Institute of Textile and Garment,Changzhou 213164,Jiangsu)

机构地区：[1]常州纺织服装职业技术学院信息服务中心,江苏常州213164

出　　处：《电脑与电信》2020年第12期53-57,共5页Computer & Telecommunication

基　　金：常州纺织服装职业技术学院2020年校级应用课题《基于IPV6技术的智慧校园多媒体教室远程集中控制系统的开发与设计》,项目编号:CFK202008。

摘　　要：近年来深度学习理论再度中兴,在机器学习视觉识别和听觉识别领域应用日益广泛。玻尔兹曼机是比较典型的深度学习神经网络,其网络权值的训练算法有多种,比较经典的如对比离差(CD)算法。目前的算法无法精确取得网络热平衡状态的期望值,只能计算近似梯度值,同时算法运算量大、运行时间长。对此,提出了一种RBM权值计算的方法。首先将RBM等效成Hopfield网络,然后利用DHNN权值设计方法设计权值矩阵,最后将RBM权值求解问题转化为求DHNN权值矩阵特征值和特征向量问题,通过实例说明计算过程并给予数据反向验证该算法的正确性。In recent years,the theory of deep learning is booming again.It is widely used in machine learning,visual recognition and auditory recognition.Boltzmann machine is a typical deep learning neural network.There are many training algorithms for its network weights,such as contrast dispersion(CD)algorithm,which is classical.However,the current algorithm cannot accurately obtain the expected value of network thermal equilibrium state.Only approximate gradient values can be calculated.At the same time,the algorithm has a large amount of computation and a long running time.In this paper,a method of RBM weight calculation is proposed.Firstly,RBM is equivalent to Hopfield network.Then the weight matrix is designed by DHNN weight design method.Finally,the RBM weight solving problem is transformed into the eigenvalue and eigenvector problem of DHNN weight matrix.An example is given to illustrate the calculation process and the correctness of the algorithm is verified by the data.

关 键 词：玻尔兹曼机 HOPFIELD网络 神经元 权值矩阵 

分 类 号：TP391.3[自动化与计算机技术—计算机应用技术]