{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,22]],"date-time":"2026-03-22T06:45:50Z","timestamp":1774161950024,"version":"3.50.1"},"reference-count":56,"publisher":"Wiley","issue":"1","license":[{"start":{"date-parts":[[2019,3,26]],"date-time":"2019-03-26T00:00:00Z","timestamp":1553558400000},"content-version":"vor","delay-in-days":84,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100003176","name":"Ministerio de Educaci\u00f3n, Cultura y Deporte","doi-asserted-by":"publisher","award":["FPU13\/03917"],"award-info":[{"award-number":["FPU13\/03917"]}],"id":[{"id":"10.13039\/501100003176","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Complexity"],"published-print":{"date-parts":[[2019,1]]},"abstract":"<jats:p>Neuroevolution is the field of study that uses evolutionary computation in order to optimize certain aspect of the design of neural networks, most often its topology and hyperparameters. The field was introduced in the late\u20101980s, but only in the latest years the field has become mature enough to enable the optimization of deep learning models, such as convolutional neural networks. In this paper, we rely on previous work to apply neuroevolution in order to optimize the topology of deep neural networks that can be used to solve the problem of handwritten character recognition. Moreover, we take advantage of the fact that evolutionary algorithms optimize a population of candidate solutions, by combining a set of the best evolved models resulting in a committee of convolutional neural networks. This process is enhanced by using specific mechanisms to preserve the diversity of the population. Additionally, in this paper, we address one of the disadvantages of neuroevolution: the process is very expensive in terms of computational time. To lessen this issue, we explore the performance of topology transfer learning: whether the best topology obtained using neuroevolution for a certain domain can be successfully applied to a different domain. By doing so, the expensive process of neuroevolution can be reused to tackle different problems, turning it into a more appealing approach for optimizing the design of neural networks topologies. After evaluating our proposal, results show that both the use of neuroevolved committees and the application of topology transfer learning are successful: committees of convolutional neural networks are able to improve classification results when compared to single models, and topologies learned for one problem can be reused for a different problem and data with a good performance. Additionally, both approaches can be combined by building committees of transferred topologies, and this combination attains results that combine the best of both approaches.<\/jats:p>","DOI":"10.1155\/2019\/2952304","type":"journal-article","created":{"date-parts":[[2019,3,26]],"date-time":"2019-03-26T23:35:45Z","timestamp":1553643345000},"update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":32,"title":["Hybridizing Evolutionary Computation and Deep Neural Networks: An Approach to Handwriting Recognition Using Committees and Transfer Learning"],"prefix":"10.1155","volume":"2019","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-8906-7572","authenticated-orcid":false,"given":"Alejandro","family":"Baldominos","sequence":"first","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0998-2907","authenticated-orcid":false,"given":"Yago","family":"Saez","sequence":"additional","affiliation":[]},{"given":"Pedro","family":"Isasi","sequence":"additional","affiliation":[]}],"member":"311","published-online":{"date-parts":[[2019,3,26]]},"reference":[{"key":"e_1_2_9_1_2","doi-asserted-by":"crossref","unstructured":"HeK. ZhangX. RenS. andSunJ. Deep residual learning for image recognition Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR \u203216) June 2016 Las Vegas Nev USA 770\u2013778 https:\/\/doi.org\/10.1109\/CVPR.2016.90.","DOI":"10.1109\/CVPR.2016.90"},{"key":"e_1_2_9_2_2","doi-asserted-by":"crossref","unstructured":"LongJ. ShelhamerE. andDarrellT. Fully convolutional networks for semantic segmentation Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR \u203215) June 2015 Boston Mass USA IEEE 3431\u20133440 https:\/\/doi.org\/10.1109\/cvpr.2015.7298965.","DOI":"10.1109\/CVPR.2015.7298965"},{"key":"e_1_2_9_3_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.neucom.2017.12.049"},{"key":"e_1_2_9_4_2","doi-asserted-by":"publisher","DOI":"10.1109\/5.726791"},{"key":"e_1_2_9_5_2","first-page":"255","volume-title":"The Handbook of Brain Theory and Neural Network","author":"LeCun Y.","year":"1998"},{"key":"e_1_2_9_6_2","doi-asserted-by":"crossref","unstructured":"BergstraJ. BreuleuxO. BastienF. LamblinP. PascanuR. DesjardinsG. TurianJ. Warde-FarleyD. andBengioY. Theano: a CPU and GPU math compiler in Python Proceedings of the 9th Python in Science Conference 2010.","DOI":"10.25080\/Majora-92bf1922-003"},{"key":"e_1_2_9_7_2","doi-asserted-by":"crossref","unstructured":"JiaY. ShelhamerE. DonahueJ. KarayevS. LongJ. GirshickR. GuadarramaS. andDarrellT. Caffe: convolutional architecture for fast feature embedding Proceedings of the ACM Conference on Multimedia (MM \u203214) November 2014 Orlando Fla USA ACM 675\u2013678 https:\/\/doi.org\/10.1145\/2647868.2654889 2-s2.0-84913580146.","DOI":"10.1145\/2647868.2654889"},{"key":"e_1_2_9_8_2","unstructured":"AbadiM. BarhamP. ChenJ. ChenZ. DavisA. DeanJ. DevinM. GhemawatS. IrvingG. IsardM. KudlurM. LevenbergJ. MongaR. MooreS. MurrayD. G. SteinerB. TuckerP. VasudevanV. WardenP. WickeM. YuY. andZhengX. TensorFlow: a system for large-scale machine learning Proceedings of the 12th USENIX Symposium on Operating Systems Design and Implementation 2016 265\u2013283."},{"key":"e_1_2_9_9_2","unstructured":"Lasagne Welcome to lasagne last visited on 2017 http:\/\/lasagne.readthedocs.io."},{"key":"e_1_2_9_10_2","unstructured":"Keras Keras: the python deep learning library 2017 https:\/\/keras.io."},{"key":"e_1_2_9_11_2","doi-asserted-by":"publisher","DOI":"10.1038\/323533a0"},{"key":"e_1_2_9_12_2","doi-asserted-by":"publisher","DOI":"10.1038\/s42256-018-0006-z"},{"key":"e_1_2_9_13_2","doi-asserted-by":"crossref","unstructured":"Koutn\u00edkJ. SchmidhuberJ. andGomezF. Evolving deep unsupervised convolutional networks for vision-based reinforcement learning Proceedings of the 2014 Annual Conference on Genetic and Evolutionary Computation July 2014 Canada 541\u2013548 2-s2.0-84905695541.","DOI":"10.1145\/2576768.2598358"},{"key":"e_1_2_9_14_2","doi-asserted-by":"crossref","unstructured":"VerbancsicsP.andHarguessJ. Image classification using generative neuroevolution for deep learning Proceedings of the 2015 IEEE Winter Conference on Applications of Computer Vision January 2015 USA 488\u2013493 2-s2.0-84925423571.","DOI":"10.1109\/WACV.2015.71"},{"key":"e_1_2_9_15_2","doi-asserted-by":"publisher","DOI":"10.1162\/artl.2009.15.2.15202"},{"key":"e_1_2_9_16_2","doi-asserted-by":"publisher","DOI":"10.1145\/2834892.2834896"},{"key":"e_1_2_9_17_2","doi-asserted-by":"crossref","unstructured":"YoungS. R. RoseD. C. JohnstonT. HellerW. T. KarnowskiT. P. PotokT. E. PattonR. M. PerdueG. andMillerJ. Evolving deep networks using HPC Proceedings of the Machine Learning on HPC Environments November 2017 Denver CO USA 3924\u20133928 https:\/\/doi.org\/10.1145\/3146347.3146355.","DOI":"10.1145\/3146347.3146355"},{"key":"e_1_2_9_18_2","unstructured":"LoshchilovI.andHutterF. CMA-ES for hyperparameter optimization of deep neural networks 2016 https:\/\/arxiv.org\/abs\/1604.07269."},{"key":"e_1_2_9_19_2","doi-asserted-by":"crossref","unstructured":"FernandoC. BanarseD. ReynoldsM. BesseF. PfauD. JaderbergM. LanctotM. andWierstraD. Convolution by evolution: differentiable pattern producing networks Proceedings of the Genetic and Evolutionary Computation Conference 2016 New York NY USA ACM 109\u2013116 MR3679650.","DOI":"10.1145\/2908812.2908890"},{"key":"e_1_2_9_20_2","unstructured":"XieL.andYuilleA. Genetic CNN Proceedings of the 16th IEEE International Conference on Computer Vision (ICCV \u203217) October 2017 2-s2.0-85041918559."},{"key":"e_1_2_9_21_2","unstructured":"MiikkulainenR. LiangJ. MeyersonE. RawalA. FinkD. FranconO. RajuB. ShahrzadH. NavruzyanA. DuffyN. andHodjatB. Evolving deep neural networks 2017 https:\/\/arxiv.org\/abs\/1703.00548."},{"key":"e_1_2_9_22_2","doi-asserted-by":"crossref","unstructured":"DesellT. Large scale evolution of convolutional neural networks using volunteer computing Proceedings of the the Genetic and Evolutionary Computation Conference Companion July 2017 Berlin Germany 127\u2013128 https:\/\/doi.org\/10.1145\/3067695.3076002.","DOI":"10.1145\/3067695.3076002"},{"key":"e_1_2_9_23_2","unstructured":"RealE. MooreS. SelleA. SaxenaS. Leon-SuematsuY. TanJ. LeQ. V. andKurakinA. Large-scale evolution of image classifiers 70 Proceedings of Machine Learning Research 2017 2902\u20132911."},{"key":"e_1_2_9_24_2","unstructured":"SunY. XueB. andZhangM. Evolving deep convolutional neural networks for image classification 2017 https:\/\/arxiv.org\/abs\/1710.10741."},{"key":"e_1_2_9_25_2","doi-asserted-by":"crossref","unstructured":"DufourqE.andBassettB. A. EDEN: evolutionary deep networks for efficient machine learning Proceedings of the Pattern Recognition Association of South Africa and Robotics and Mechatronics (PRASA-RobMech) November 2017 Bloemfontein South Africa 110\u2013115 https:\/\/doi.org\/10.1109\/RoboMech.2017.8261132.","DOI":"10.1109\/RoboMech.2017.8261132"},{"key":"e_1_2_9_26_2","unstructured":"SuganumaM. ShirakawaS. andNagaoT. A genetic programming approach to designing convolutional neural network architectures Proceedings of the Genetic and Evolutionary Computation Conference (GECCO \u203217) July 2017 Germany 497\u2013504 2-s2.0-85026354551."},{"key":"e_1_2_9_27_2","unstructured":"DavisonJ. DEvol: Automated deep neural network design via genetic programming 2017 https:\/\/github.com\/joeddav\/devol."},{"key":"e_1_2_9_28_2","unstructured":"BochinskiE. SenstT. andSikoraT. Hyper-parameter optimization for convolutional neural network committees based on evolutionary algorithms Proceedings of the 24th IEEE International Conference on Image Processing (ICIP \u203217) September 2017 Beijing China 3924\u20133928 2-s2.0-85045336519."},{"key":"e_1_2_9_29_2","unstructured":"LiuH. SimonyanK. VinyalsO. FernandoC. andKavukcuogluK. Hierarchical representations for efficient architecture search Proceedings of the 6th International Conference on Learning Representations 2018 Vancouver British Columbia Canada."},{"key":"e_1_2_9_30_2","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-77538-8_27"},{"key":"e_1_2_9_31_2","doi-asserted-by":"crossref","unstructured":"PrellbergJ.andKramerO. Lamarckian evolution of convolutional neural networks 2018 https:\/\/arxiv.org\/abs\/1806.08099 https:\/\/doi.org\/10.1109\/IJCNN.2018.8489647.","DOI":"10.1109\/IJCNN.2018.8489647"},{"key":"e_1_2_9_32_2","article-title":"DENSER: deep evolutionary network structured representation","author":"Assun\u00e7\u00e3o F.","year":"2018","journal-title":"Genetic Programming and Evolvable Machines"},{"key":"e_1_2_9_33_2","unstructured":"WangB. SunY. XueB. andZhangM. A hybrid DE approach to designing CNN for image classification Proceedings of the 31st Australasian Joint Conference on Artificial Intelligence 2018."},{"key":"e_1_2_9_34_2","unstructured":"SunY. XueB. andZhangM. Automatically evolving cnn architectures based on blocks 2018 https:\/\/arxiv.org\/abs\/1810.11875."},{"key":"e_1_2_9_35_2","unstructured":"Cire\u015fanD. C. MeierU. GambardellaL. M. andSchmidhuberJ. Convolutional neural network committees for handwritten character classification Proceedings of the 11th International Conference on Document Analysis and Recognition September 2011 1135\u20131139 2-s2.0-82355186196."},{"key":"e_1_2_9_36_2","doi-asserted-by":"publisher","DOI":"10.1109\/TAFFC.2017.2753235"},{"key":"e_1_2_9_37_2","doi-asserted-by":"publisher","DOI":"10.1051\/0004-6361\/201731201"},{"key":"e_1_2_9_38_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.cviu.2017.12.005"},{"key":"e_1_2_9_39_2","doi-asserted-by":"publisher","DOI":"10.1109\/JBHI.2016.2635663"},{"key":"e_1_2_9_40_2","unstructured":"BakerB. GuptaO. NaikN. andRaskarR. Designing neural network architectures using reinforcement learning Proceedings of the 5th International Conference on Learning Representations 2017."},{"key":"e_1_2_9_41_2","doi-asserted-by":"publisher","DOI":"10.1109\/tkde.2009.191"},{"key":"e_1_2_9_42_2","article-title":"Batch-normalized maxout network in network","volume":"48","author":"Chang J. R.","year":"2015","journal-title":"Journal of Machine Learning Research"},{"key":"e_1_2_9_43_2","doi-asserted-by":"crossref","unstructured":"CohenG. AfsharS. TapsonJ. andvan SchaikA. EMNIST: an extension of MNIST to handwritten letters 2017 https:\/\/arxiv.org\/abs\/1702.05373.","DOI":"10.1109\/IJCNN.2017.7966217"},{"key":"e_1_2_9_44_2","unstructured":"NIST NIST Special Database 19 2017 https:\/\/www.nist.gov\/srd\/nist-special-database-19."},{"key":"e_1_2_9_45_2","unstructured":"GrotherP. J.andHanaokaK. K. NIST special database 19 handprinted forms and characters database 2016 National Institute of Standards and Technology."},{"key":"e_1_2_9_46_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.neucom.2014.01.071"},{"key":"e_1_2_9_47_2","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-68935-7_42"},{"key":"e_1_2_9_48_2","doi-asserted-by":"crossref","unstructured":"SinghS. PaulA. andArunM. Parallelization of digit recognition system using deep convolutional neural network on CUDA Proceedings of the Third International Conference on Sensing Signal Processing and Security (ICSSS \u203217) May 2017 Chennai India 379\u2013383 https:\/\/doi.org\/10.1109\/SSPS.2017.8071623.","DOI":"10.1109\/SSPS.2017.8071623"},{"key":"e_1_2_9_49_2","doi-asserted-by":"crossref","DOI":"10.3389\/fnins.2017.00324","article-title":"Event-driven random back-propagation: enabling neuromorphic deep learning machines","volume":"11","author":"Neftci E. O.","year":"2017","journal-title":"Frontiers in Neuroscience"},{"key":"e_1_2_9_50_2","unstructured":"ShuL. XuH. andLiuB. Unseen class discovery in open-world classification 2018 https:\/\/arxiv.org\/abs\/1801.05609."},{"key":"e_1_2_9_51_2","doi-asserted-by":"crossref","unstructured":"MilgramJ. CherietM. andSabourinR. Estimating accurate multi-class probabilities with support vector machines Proceedings of the International Joint Conference on Neural Networks (IJCNN \u203205) August 2005 Canada 1906\u20131911 2-s2.0-33750104507.","DOI":"10.1109\/IJCNN.2005.1556171"},{"key":"e_1_2_9_52_2","doi-asserted-by":"publisher","DOI":"10.13176\/11.23"},{"key":"e_1_2_9_53_2","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2002.1046154"},{"key":"e_1_2_9_54_2","unstructured":"RadtkeP. V. W. SabourinR. andWongT. Using the RRT algorithm to optimize classification systems for handwritten digits and letters Proceedings of the 23rd Annual ACM Symposium on Applied Computing (SAC \u203208) March 2008 Brazil 1748\u20131752 2-s2.0-56749171135."},{"key":"e_1_2_9_55_2","unstructured":"KoerichA. L.andKalvaP. R. Unconstrained handwritten character recognition using metaclasses of characters Proceedings of the IEEE International Conference on Image Processing (ICIP \u203205) September 2005 Italy 542\u2013545 2-s2.0-33749620053."},{"key":"e_1_2_9_56_2","doi-asserted-by":"crossref","unstructured":"CavalinP. R. De Souza BrittoA.Jr. BortolozziF. SabourinR. andOliveiraL. E. S. An implicit segmentation-based method for recognition of handwritten strings of characters Proceedings of the 2006 ACM Symposium on Applied Computing April 2006 France 836\u2013840 2-s2.0-33751026721.","DOI":"10.1145\/1141277.1141468"}],"container-title":["Complexity"],"original-title":[],"language":"en","link":[{"URL":"http:\/\/downloads.hindawi.com\/journals\/complexity\/2019\/2952304.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"http:\/\/downloads.hindawi.com\/journals\/complexity\/2019\/2952304.xml","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1155\/2019\/2952304","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,8,7]],"date-time":"2024-08-07T14:33:12Z","timestamp":1723041192000},"score":1,"resource":{"primary":{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/10.1155\/2019\/2952304"}},"subtitle":[],"editor":[{"given":"Michele","family":"Scarpiniti","sequence":"additional","affiliation":[]}],"short-title":[],"issued":{"date-parts":[[2019,1]]},"references-count":56,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2019,1]]}},"alternative-id":["10.1155\/2019\/2952304"],"URL":"https:\/\/doi.org\/10.1155\/2019\/2952304","archive":["Portico"],"relation":{},"ISSN":["1076-2787","1099-0526"],"issn-type":[{"value":"1076-2787","type":"print"},{"value":"1099-0526","type":"electronic"}],"subject":[],"published":{"date-parts":[[2019,1]]},"assertion":[{"value":"2018-12-03","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2019-03-03","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2019-03-26","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"2952304"}}