A Novel Two-stage Learning Pipeline for Deep Neural Networks

被引:3
|
作者
Ding, Chunhui [1 ]
Hu, Zhengwei [1 ]
Karmoshi, Saleem [1 ]
Zhu, Ming [1 ]
机构
[1] Univ Sci & Technol China, Hefei, Anhui, Peoples R China
关键词
DNN; Two-stage structure; Big data;
D O I
10.1007/s11063-017-9578-6
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
In this work, a training method was proposed for Deep Neural Networks (DNNs) based on a two-stage structure. Local DNN models are trained in all local machines and uploaded to the center with partial training data. These local models are integrated as a new DNN model (combination DNN). With another DNN model (optimization DNN) connected, the combination DNN forms a global DNN model in the center. This results in greater accuracy than local DNN models with smaller amounts of data uploaded. In this case, the bandwidth of the uploaded data is saved, and the accuracy is maintained as well. Experiments are conducted on MNIST dataset, CIFAR-10 dataset and LFW dataset. The results show that with less training data uploaded, the global model produces greater accuracy than local models. Specifically, this method focuses on condition of big data.
引用
收藏
页码:159 / 169
页数:11
相关论文
共 50 条
  • [21] A Novel Two-Stage Deep Learning Model for Efficient Network Intrusion Detection
    Khan, Farrukh Aslam
    Gumaei, Abdu
    Derhab, Abdelouahid
    Hussain, Amir
    IEEE ACCESS, 2019, 7 : 30373 - 30385
  • [22] Accurate identification and measurement of the precipitate area by two-stage deep neural networks in novel chromium-based alloys
    Xia, Zeyu
    Ma, Kan
    Cheng, Sibo
    Blackburn, Thomas
    Peng, Ziling
    Zhu, Kewei
    Zhang, Weihang
    Xiao, Dunhui
    Knowles, Alexander J.
    Arcucci, Rossella
    PHYSICAL CHEMISTRY CHEMICAL PHYSICS, 2023, 25 (23) : 15970 - 15987
  • [23] Two-Stage Hybrid Approach of Deep Learning Networks for Interstitial Lung Disease Classification
    Pawar, Swati P.
    Talbar, Sanjay N.
    BIOMED RESEARCH INTERNATIONAL, 2022, 2022
  • [24] A two-stage strategy for brain-inspired unsupervised learning in spiking neural networks
    Cao, Zhen
    Ma, Chuanfeng
    Hou, Biao
    Chen, Xiaoyu
    Li, Leida
    Zhu, Hao
    Quan, Dou
    Jiao, Licheng
    NEUROCOMPUTING, 2025, 611
  • [25] Lung Nodule Classification using A Novel Two-stage Convolutional Neural Networks Structure
    An, Yang
    Hu, Tianren
    Wang, Jiaqi
    Lyu, Juan
    Banerjee, Sunctra
    Ling, Sai Ho
    2019 41ST ANNUAL INTERNATIONAL CONFERENCE OF THE IEEE ENGINEERING IN MEDICINE AND BIOLOGY SOCIETY (EMBC), 2019, : 6259 - 6262
  • [26] Two-Stage Monaural Source Separation in Reverberant Room Environments Using Deep Neural Networks
    Sun, Yang
    Wang, Wenwu
    Chambers, Jonathon
    Naqvi, Syed Mohsen
    IEEE-ACM TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSING, 2019, 27 (01) : 125 - 139
  • [27] A two-stage data quality improvement strategy for deep neural networks in fault severity estimation
    Yao, Yuan
    Wu, Lan
    Xie, Bin
    Lei, Li
    Wang, Zaixiang
    Li, Yesong
    MECHANICAL SYSTEMS AND SIGNAL PROCESSING, 2023, 200
  • [28] A Two-Stage Approach for the Remaining Useful Life Prediction of Bearings Using Deep Neural Networks
    Xia, Min
    Li, Teng
    Shu, Tongxin
    Wan, Jiafu
    de Silva, Clarence W.
    Wang, Zhongren
    IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS, 2019, 15 (06) : 3703 - 3711
  • [29] Two-Stage Single-Channel Audio Source Separation Using Deep Neural Networks
    Grais, Emad M.
    Roma, Gerard
    Simpson, Andrew J. R.
    Plumbley, Mark D.
    IEEE-ACM TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSING, 2017, 25 (09) : 1469 - 1479
  • [30] A two-stage learning framework of relational Markov networks
    Wan, Huaiyu
    Lin, Youfang
    Wu, Zhihao
    Huang, Houkuan
    Journal of Computational Information Systems, 2010, 6 (04): : 1027 - 1035