Gradient descent learning algorithm for hierarchical neural networks: A case study in industrial quality

被引:0
|
作者
Baratta, D [1 ]
Diotalevi, F [1 ]
Valle, M [1 ]
Caviglia, DD [1 ]
机构
[1] Univ Genoa, Dept Biophys & Elect Engn, I-16145 Genoa, Italy
关键词
D O I
暂无
中图分类号
TP301 [理论、方法];
学科分类号
081202 ;
摘要
This paper deals with the training procedure for a hierarchical neural network (Tree of Multi-Layer Perceptrons - TMLP) aimed to classify surface defects in flat rolled strips. Due to the difficulties in collecting large Data Bases it is necessary to exploit at the best the available knowledge. A comparison between techniques derived from both the Back-Propagation and Weight-Perturbation algorithms is done, and experimental results are reported.
引用
收藏
页码:578 / 587
页数:10
相关论文
共 50 条
  • [31] Learning Overparameterized Neural Networks via Stochastic Gradient Descent on Structured Data
    Li, Yuanzhi
    Liang, Yingyu
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 31 (NIPS 2018), 2018, 31
  • [32] Crossprop: Learning Representations by Stochastic Meta-Gradient Descent in Neural Networks
    Veeriah, Vivek
    Zhang, Shangtong
    Sutton, Richard S.
    MACHINE LEARNING AND KNOWLEDGE DISCOVERY IN DATABASES, ECML PKDD 2017, PT I, 2017, 10534 : 445 - 459
  • [33] LEARNING SHALLOW NEURAL NETWORKS VIA PROVABLE GRADIENT DESCENT WITH RANDOM INITIALIZATION
    Xia, Shuhao
    Shi, Yuanming
    2019 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP), 2019, : 5616 - 5620
  • [34] The Dynamics of Gradient Descent for Overparametrized Neural Networks
    Satpathi, Siddhartha
    Srikant, R.
    LEARNING FOR DYNAMICS AND CONTROL, VOL 144, 2021, 144
  • [35] Applying Gradient Descent in Convolutional Neural Networks
    Cui, Nan
    2ND INTERNATIONAL CONFERENCE ON MACHINE VISION AND INFORMATION TECHNOLOGY (CMVIT 2018), 2018, 1004
  • [36] A Fast Adaptive Online Gradient Descent Algorithm in Over-Parameterized Neural Networks
    Yang, Anni
    Li, Dequan
    Li, Guangxiang
    NEURAL PROCESSING LETTERS, 2023, 55 (04) : 4641 - 4659
  • [37] A Fast Adaptive Online Gradient Descent Algorithm in Over-Parameterized Neural Networks
    Anni Yang
    Dequan Li
    Guangxiang Li
    Neural Processing Letters, 2023, 55 : 4641 - 4659
  • [38] Adaptive orthogonal gradient descent algorithm for fully complex-valued neural networks
    Zhao, Weijing
    Huang, He
    NEUROCOMPUTING, 2023, 546
  • [39] Learning rates of gradient descent algorithm for classification
    Dong, Xue-Mei
    Chen, Di-Rong
    JOURNAL OF COMPUTATIONAL AND APPLIED MATHEMATICS, 2009, 224 (01) : 182 - 192
  • [40] Artificial Neural Networks with Gradient Learning Algorithm for Credit Scoring
    Soydaner, Derya
    Kocadagli, Ozan
    ISTANBUL UNIVERSITY JOURNAL OF THE SCHOOL OF BUSINESS, 2015, 44 (02): : 3 - 12