L1正则化的深度谱聚类算法  

Deep spectral clustering algorithm with L1 regularization

作　　者：李文博[1,2] 刘波[1,2] 陶玲玲 罗棻 张航[1,2] LI Wenbo;LIU Bo;TAO Lingling;LUO Fen;ZHANG Hang(School of Artificial Intelligence,Chongqing Technology and Business University,Chongqing 400067,China;Chongqing Key Laboratory of Intelligent Perception and Block Chain Technology(Chongqing Technology and Business University),Chongqing 400067,China)

机构地区：[1]重庆工商大学人工智能学院,重庆400067 [2]智能感知与区块链技术重庆市重点实验室(重庆工商大学),重庆400067

出　　处：《计算机应用》2023年第12期3662-3667,共6页journal of Computer Applications

基　　金：重庆市教委科学技术研究项目(KJZD-K202200803);重庆市自然科学基金资助项目(cstc2018jcyjAX0057);重庆工商大学研究生“创新型科研项目”(yjscxx2022-112-68)。

摘　　要：针对深度谱聚类模型训练不稳定和泛化能力弱等问题,提出L1正则化的深度谱聚类算法(DSCLR)。首先,在深度谱聚类的目标函数中引入L1正则化,使深度神经网络模型生成的拉普拉斯矩阵的特征向量稀疏化,并提升模型的泛化能力;其次,通过利用参数化修正线性单元激活函数(PReLU)改进基于深度神经网络的谱聚类算法的网络结构,解决模型训练不稳定和欠拟合问题。在MNIST数据集上的实验结果表明,所提算法在聚类精度(CA)、归一化互信息(NMI)指数和调整兰德系数(ARI)这3个评价指标上,相较于深度谱聚类算法分别提升了11.85、7.75和17.19个百分点。此外,所提算法相较于深度嵌入聚类(DEC)和基于对偶自编码器网络的深度谱聚类(DSCDAN)等算法,在CA、NMI和ARI这3个评价指标上也有大幅提升。Aiming at the problems that the deep spectral clustering models perform poorly in training stability and generalization capability,a Deep Spectral Clustering algorithm with L1 Regularization(DSCLR)was proposed.Firstly,L1 regularization was introduced into the objective function of deep spectral clustering to sparsify the eigen vectors of the Laplacian matrix generated by the deep neural network model.And the generalization capability of the model was enhanced.Secondly,the network structure of the spectral clustering algorithm based on deep neural network was improved by using the Parametric Rectified Linear Unit activation function(PReLU)to solve the problems of model training instability and underfitting.Experimental results on MNIST dataset show that the proposed algorithm improves Clustering Accuracy(CA),Normalized Mutual Information(NMI)index,and Adjusted Rand Index(ARI)by 11.85,7.75,and 17.19 percentage points compared to the deep spectral clustering algorithm,respectively.Furthermore,the proposed algorithm also significantly improves the three evaluation metrics,CA,NMI and ARI,compared to algorithms such as Deep Embedded Clustering(DEC)and Deep Spectral Clustering using Dual Autoencoder Network(DSCDAN).

关 键 词：深度聚类 谱聚类 L1正则化 深度学习 无监督学习 

分 类 号：TP311.5[自动化与计算机技术—计算机软件与理论]