BP Neural Network Feature Selection Based on Group Lasso Regularization

被引：0
|
作者：
Liu, Tiqian [1 ]
Xiao, Jiang-Wen [1 ]
Huang, Zhengyi [1 ]
Kong, Erdan [1 ]
Liang, Yuntao [1 ]
机构：
[1] Huazhong Univ Sci & Technol, Sch Artificial Intelligence & Automat, Wuhan, Peoples R China
来源：
2019 CHINESE AUTOMATION CONGRESS (CAC2019) | 2019年
基金：
中国国家自然科学基金;
关键词：
group lasso; feature selection; BP neural network; feature weight coefficient;
D O I：
10.1109/cac48633.2019.8996679
中图分类号：
TP [自动化技术、计算机技术];
学科分类号：
0812 ;
摘要：
In this paper, the feature weight coefficient of neural network is proposed based on the results of neural network feature selection. Firstly, this paper introduces the neural network feature selection method based on Group Least absolute shrinkage and selection operator (Lasso) penalty, smooth approximation technique and the feature weight coefficient. Then, a two-stage algorithm for back propagation (BP) neural network feature selection based on Group Lasso penalty term is proposed. And then, the paper compares and analyzes the influence of smooth approximation technique on BP neural network feature selection. Finally, the reliability of the algorithm and the validity of the feature weight coefficient are verified by comparing the results of four datasets.
引用
收藏
页码：2786 / 2790
页数：5
相关论文
共 50 条
  • [41] Feature Compression based BP Neural Network for IoT Performance Prediction
    Zhao, Ziru
    Xu, Yanhong
    Li, Zhao
    Chang, Zhixian
    Liu, Jia
    2023 IEEE 9TH WORLD FORUM ON INTERNET OF THINGS, WF-IOT, 2023,
  • [42] An algorithm of texture classification based on feature extraction and BP neural network
    Liu, Tongyan
    Liu, Zongguo
    Wu, Guoqing
    Journal of Information and Computational Science, 2015, 12 (06): : 2315 - 2323
  • [43] Group-Lasso Regularized Artificial Neural Network for Embedded Feature Selection in Radiomics Study On Pediatric Patients with Craniopharyngioma Treated with Proton Therapy
    Yang, W.
    Hua, C.
    Davis, T.
    Uh, J.
    Merchant, T.
    MEDICAL PHYSICS, 2021, 48 (06)
  • [44] Convergence analyses on sparse feedforward neural networks via group lasso regularization
    Wang, Jian
    Cai, Qingling
    Chang, Qingquan
    Zurada, Jacek M.
    INFORMATION SCIENCES, 2017, 381 : 250 - 269
  • [45] THE EFFICIENCY ANALYSIS OF THE STATISTICAL FEATURE IN NETWORK TRAFFIC IDENTIFICATION BASED ON BP NEURAL NETWORK
    Mu, Cheng
    Zhang, Changzhi
    Huang, Xiaohong
    Ma, Yan
    2013 5TH IEEE INTERNATIONAL CONFERENCE ON BROADBAND NETWORK & MULTIMEDIA TECHNOLOGY (IC-BNMT), 2013, : 70 - 74
  • [46] Group Lasso Regularized Multiple Kernel Learning for Heterogeneous Feature Selection
    Yeh, Yi-Ren
    Chung, Yung-Yu
    Lin, Ting-Chu
    Wang, Yu-Chiang Frank
    2011 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS (IJCNN), 2011, : 2570 - 2577
  • [47] Deep Feature Selection using an Enhanced Sparse Group Lasso Algorithm
    Farokhmanesh, Fatemeh
    Sadeghi, Mohammad Taghi
    2019 27TH IRANIAN CONFERENCE ON ELECTRICAL ENGINEERING (ICEE 2019), 2019, : 1549 - 1552
  • [48] Convolutional Neural Network Simplification based on Feature Maps Selection
    Rui, Ting
    Zou, Junhua
    Zhou, You
    Fei, Jianchao
    Yang, Chengsong
    2016 IEEE 22ND INTERNATIONAL CONFERENCE ON PARALLEL AND DISTRIBUTED SYSTEMS (ICPADS), 2016, : 1207 - 1210
  • [49] A Joint Evolutionary Method Based on Neural Network For Feature Selection
    Zhang, Biying
    ICICTA: 2009 SECOND INTERNATIONAL CONFERENCE ON INTELLIGENT COMPUTATION TECHNOLOGY AND AUTOMATION, VOL I, PROCEEDINGS, 2009, : 7 - 10
  • [50] Convolutional neural network feature maps selection based on LDA
    Rui, Ting
    Zou, Junhua
    Zhou, You
    Fei, Jianchao
    Yang, Chengsong
    MULTIMEDIA TOOLS AND APPLICATIONS, 2018, 77 (09) : 10635 - 10649
12345