{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,27]],"date-time":"2026-09-27T03:46:09Z","timestamp":1790480769500,"version":"4.1.0"},"reference-count":28,"publisher":"MDPI AG","issue":"3","license":[{"start":{"date-parts":[[2018,3,7]],"date-time":"2018-03-07T00:00:00Z","timestamp":1520380800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["61640209 and 91746116"],"award-info":[{"award-number":["61640209 and 91746116"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Foundation for Distinguished Young Talents of Guizhou Province","award":["QKHRZ[2015]13"],"award-info":[{"award-number":["QKHRZ[2015]13"]}]},{"DOI":"10.13039\/501100004001","name":"Science and Technology Foundation of Guizhou Province","doi-asserted-by":"publisher","award":["JZ[2014]2004, JZ[2014]2001, ZDZX[2013]6020, and LH[2016]7433"],"award-info":[{"award-number":["JZ[2014]2004, JZ[2014]2001, ZDZX[2013]6020, and LH[2016]7433"]}],"id":[{"id":"10.13039\/501100004001","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Basic Platform Program of Guizhou Province","award":["QKHPTRC[2018]5702 and [2017]5788"],"award-info":[{"award-number":["QKHPTRC[2018]5702 and [2017]5788"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Algorithms"],"abstract":"<jats:p>This study proposes a modified convolutional neural network (CNN) algorithm that is based on dropout and the stochastic gradient descent (SGD) optimizer (MCNN-DS), after analyzing the problems of CNNs in extracting the convolution features, to improve the feature recognition rate and reduce the time-cost of CNNs. The MCNN-DS has a quadratic CNN structure and adopts the rectified linear unit as the activation function to avoid the gradient problem and accelerate convergence. To address the overfitting problem, the algorithm uses an SGD optimizer, which is implemented by inserting a dropout layer into the all-connected and output layers, to minimize cross entropy. This study used the datasets MNIST, HCL2000, and EnglishHand as the benchmark data, analyzed the performance of the SGD optimizer under different learning parameters, and found that the proposed algorithm exhibited good recognition performance when the learning rate was set to [0.05, 0.07]. The performances of WCNN, MLP-CNN, SVM-ELM, and MCNN-DS were compared. Statistical results showed the following: (1) For the benchmark MNIST, the MCNN-DS exhibited a high recognition rate of 99.97%, and the time-cost of the proposed algorithm was merely 21.95% of MLP-CNN, and 10.02% of SVM-ELM; (2) Compared with SVM-ELM, the average improvement in the recognition rate of MCNN-DS was 2.35% for the benchmark HCL2000, and the time-cost of MCNN-DS was only 15.41%; (3) For the EnglishHand test set, the lowest recognition rate of the algorithm was 84.93%, the highest recognition rate was 95.29%, and the average recognition rate was 89.77%.<\/jats:p>","DOI":"10.3390\/a11030028","type":"journal-article","created":{"date-parts":[[2018,3,7]],"date-time":"2018-03-07T12:55:23Z","timestamp":1520427323000},"page":"28","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":119,"title":["Modified Convolutional Neural Network Based on Dropout and the Stochastic Gradient Descent Optimizer"],"prefix":"10.3390","volume":"11","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-1915-9487","authenticated-orcid":false,"given":"Jing","family":"Yang","sequence":"first","affiliation":[{"name":"Key Laboratory of Advanced Manufacturing Technology of Ministry of Education, Guizhou University, Jixie Building 405 of West Campus, Huaxi District, Guiyang 550025, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8761-5195","authenticated-orcid":false,"given":"Guanci","family":"Yang","sequence":"additional","affiliation":[{"name":"Key Laboratory of Advanced Manufacturing Technology of Ministry of Education, Guizhou University, Jixie Building 405 of West Campus, Huaxi District, Guiyang 550025, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2018,3,7]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"58","DOI":"10.1016\/j.neubiorev.2017.01.002","article-title":"Using deep learning to investigate the neuroimaging correlates of psychiatric and neurological disorders: Methods and applications","volume":"74","author":"Vieira","year":"2017","journal-title":"Neurosci. Biobehav. Rev."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"48","DOI":"10.1016\/j.neucom.2016.11.046","article-title":"A Fast and Memory Saved GPU Acceleration Algorithm of Convolutional Neural Networks for Target Detection","volume":"230","author":"Li","year":"2017","journal-title":"Neurocomputing"},{"key":"ref_3","first-page":"226","article-title":"GPU-based parallel optimization of immune convolutional neural network and embedded system","volume":"36","author":"Gong","year":"2016","journal-title":"Eng. Appl. Artif. Intell."},{"key":"ref_4","first-page":"4113","article-title":"Weights and structure determination method of multiple-input Sigmoid activation function neural network","volume":"29","author":"Zhang","year":"2012","journal-title":"Appl. Res. Comput."},{"key":"ref_5","doi-asserted-by":"crossref","unstructured":"Chen, L., Wu, C., Fan, W., Sun, J., and Naoi, S. (2014). Adaptive Local Receptive Field Convolutional Neural Networks for Handwritten Chinese Character Recognition. Chinese Conference on Pattern Recognition, Springer.","DOI":"10.1007\/978-3-662-45643-9_48"},{"key":"ref_6","doi-asserted-by":"crossref","unstructured":"Singh, P., Verma, A., and Chaudhari, N.S. (2016). Deep Convolutional Neural Network Classifier for Handwritten Devanagari Character Recognition. Information Systems Design and Intelligent Applications, Springer.","DOI":"10.1007\/978-81-322-2752-6_54"},{"key":"ref_7","first-page":"110","article-title":"A novel companion objective function for regularization of deepconvolutional neural networks","volume":"56","author":"Sun","year":"2016","journal-title":"Image Vis. Comput."},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Wachinger, C., Reuter, M., and Klein, T. (2017). DeepNAT: Deep convolutional neural network for segmenting neuroanatomy. NeuroImage.","DOI":"10.1016\/j.neuroimage.2017.02.035"},{"key":"ref_9","first-page":"357","article-title":"Semantic Image Segmentation with Deep Convolutional Nets and Fully Connected CRFs","volume":"26","author":"Chen","year":"2016","journal-title":"Comput. Sci."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"98","DOI":"10.3103\/S1060992X11020032","article-title":"CUDA-enabled implementation of a neural network algorithm for handwritten digit recognition","volume":"20","author":"Izotov","year":"2011","journal-title":"Opt. Mem. Neural Netw."},{"key":"ref_11","first-page":"1247","article-title":"Training sample selection method for Neural Networks based on Nearest neighbor rule","volume":"33","author":"Hao","year":"2007","journal-title":"Acta Autom. Sin."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"35","DOI":"10.1016\/j.ascom.2017.01.002","article-title":"Radio frequency interference mitigation using deep convolutional neural networks","volume":"18","author":"Akeret","year":"2017","journal-title":"Astron. Comput."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"81","DOI":"10.1016\/j.neunet.2016.06.002","article-title":"Pointwise and uniform approximation by multivariate neural network operators of the max-product type","volume":"81","author":"Costarelli","year":"2016","journal-title":"Neural Netw."},{"key":"ref_14","unstructured":"Lee, C.Y., Xie, S., Gallagher, P., Zhang, Z., and Tu, Z. (2015, January 21). Deeply-supervised nets. Proceedings of the Eighteenth International Conference on Artificial Intelligence and Statistics, San Diego, CA, USA."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"2278","DOI":"10.1109\/5.726791","article-title":"Gradient-based learning applied to document recognition","volume":"86","author":"LeCun","year":"1998","journal-title":"Proc. IEEE"},{"key":"ref_16","first-page":"1229","article-title":"Review of Convolutional Neural Network","volume":"40","author":"Zhou","year":"2017","journal-title":"Chin. J. Comput."},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"22","DOI":"10.1186\/s40537-017-0084-5","article-title":"Large-scale distributed L-BFGS","volume":"4","author":"Najafabadi","year":"2017","journal-title":"J. Big Data"},{"key":"ref_18","unstructured":"Zinkevich, M., Weimer, M., Li, L., and Smola, A.J. (2010). Parallelized stochastic gradient descent. Advances in Neural Information Processing Systems, Neural Information Processing Systems Foundation, Inc."},{"key":"ref_19","unstructured":"Hardt, M., Recht, B., and Singer, Y. (arXiv, 2015). Train faster, generalize better: Stability of stochastic gradient descent, arXiv."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"76","DOI":"10.1016\/S1004-4132(07)60055-3","article-title":"Robust fault-tolerant control for networked control system with packet dropout","volume":"18","author":"Hou","year":"2007","journal-title":"J. Syst. Eng. Electron."},{"key":"ref_21","first-page":"33","article-title":"Research on Quantum Neural Network and its Applications Based on Tanh Activation Function","volume":"16","author":"Luo","year":"2012","journal-title":"Comput. Digit. Eng."},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"37","DOI":"10.1016\/j.neucom.2017.07.061","article-title":"A joint residual network with paired ReLUs activation for image super-resolution","volume":"273","author":"Tang","year":"2018","journal-title":"Neurocomputing"},{"key":"ref_23","unstructured":"G\u00fcnnemann, N., and Pfeffer, J. (2017, January 11). Predicting Defective Engines using Convolutional Neural Networks on Temporal Vibration Signals. Proceedings of the First International Workshop on Learning with Imbalanced Domains: Theory and Applications, Munich, Germany."},{"key":"ref_24","doi-asserted-by":"crossref","unstructured":"Jin, X., Xu, C., Feng, J., Wei, Y., Xiong, J., and Yan, S. (2016, January 12\u201317). Deep learning with S-shaped rectified linear activation units. Proceedings of the Thirtieth AAAI Conference on Artificial Intelligence (AAAI Press), Phoenix, AZ, USA.","DOI":"10.1609\/aaai.v30i1.10287"},{"key":"ref_25","first-page":"167","article-title":"Image Classification Based on Mixed Deep Learning Model Transfer Learning","volume":"28","author":"Shi","year":"2016","journal-title":"J. Syst. Simul."},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"190","DOI":"10.1016\/j.patcog.2016.04.007","article-title":"DropSample: A new training method to enhance deep convolutional neural networks for large-scale unconstrained handwritten Chinese character recognition","volume":"58","author":"Yang","year":"2016","journal-title":"Pattern Recognit."},{"key":"ref_27","first-page":"107","article-title":"Reduced extreme learning machine employing SVM technique","volume":"42","author":"Shen","year":"2014","journal-title":"J. Huazhong Univ. Sci. Technol."},{"key":"ref_28","unstructured":"Yang, G.-C., Yang, J., Su, Z.D., and Chen, Z.-J. (2018). Improved YOLO feature extraction algorithm and its application to privacy situation detection of social robots. Acta Autom. Sin., 1\u201312."}],"container-title":["Algorithms"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1999-4893\/11\/3\/28\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,11]],"date-time":"2025-10-11T14:57:57Z","timestamp":1760194677000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1999-4893\/11\/3\/28"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2018,3,7]]},"references-count":28,"journal-issue":{"issue":"3","published-online":{"date-parts":[[2018,3]]}},"alternative-id":["a11030028"],"URL":"https:\/\/doi.org\/10.3390\/a11030028","relation":{},"ISSN":["1999-4893"],"issn-type":[{"value":"1999-4893","type":"electronic"}],"subject":[],"published":{"date-parts":[[2018,3,7]]}}}