A Survey of Handwritten Character Recognition with MNIST and EMNIST
Open accessThe DOI matches its Crossref record; 84 of the 84 references it cites were found in a registry or an index.
Abstract
This paper summarizes the top state-of-the-art contributions reported on the MNIST dataset for handwritten digit recognition. This dataset has been extensively used to validate novel techniques in computer vision, and in recent years, many authors have explored the performance of convolutional neural networks (CNNs) and other deep learning techniques over this dataset. To the best of our knowledge, this paper is the first exhaustive and updated review of this dataset; there are some online rankings, but they are outdated, and most published papers survey only closely related works, omitting most of the literature. This paper makes a distinction between those works using some kind of data augmentation and works using the original dataset out-of-the-box. Also, works using CNNs are reported separately; as they are becoming the state-of-the-art approach for solving this problem. Nowadays, a significant amount of works have attained a test error rate smaller than 1% on this dataset; which is becoming non-challenging. By mid-2017, a new dataset was introduced: EMNIST, which involves both digits and letters, with a larger amount of data acquired from a database different than MNIST’s. In this paper, EMNIST is explained and some results are surveyed.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Figures
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Although the images are not of a real-world environment, they can be an appropriate alternative considering that the main objective of this study was to explore the possibility of deep learning for the objective quantification of the visual perception of the substituted signal.”
From Deep Learning-Based Optimization of Visual–Auditory Sensory Substitution (2023) · cites this paper as [31]
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Gradient-based learning applied to document recognition
Lecun, Y., Bottou, L., Bengio, Y., & Haffner, P. (1998) · Proceedings of the IEEE · doi:10.1109/5.726791 · Cited by 58,805 · cited 7×
“For instance, Belongie et al. [5] achieved 0.63% and Keysers et al. [6] achieved 0.54% and 0.52% using K-NN, Kégl and Busa-Fekete [7] achieved 0.87% using boosted stumps on Haar features, LeCun et al. [4] achieved 0.8% and Decoste and Schölkopf [8] attained results from 0.56 to 0.68% using SVM.”
BackgroundMethodInfluentialExtreme learning machine: Theory and applications
Huang, G.-B., Zhu, Q.-Y., & Siew, C.-K. (2006) · Neurocomputing · doi:10.1016/j.neucom.2005.12.126 · Cited by 13,285 · cited 1×
“In 2015, McDonnell et al. [21] proposed an approach using the ‘extreme learning machine’ (ELM) [22] algorithm to train shallow non-convolutional neural networks, and they agree on the fact that data augmentation with distortions can reduce error rates even further.”
MethodShape matching and object recognition using shape contexts
Belongie, S., Malik, J., & Puzicha, J. (2002) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/34.993558 · Cited by 6,306 · cited 2×
“For instance, Belongie et al. [5] achieved 0.63% and Keysers et al. [6] achieved 0.54% and 0.52% using K-NN, Kégl and Busa-Fekete [7] achieved 0.87% using boosted stumps on Haar features, LeCun et al. [4] achieved 0.8% and Decoste and Schölkopf [8] attained results from 0.56 to 0.68% using SVM.”
MethodInfluentialDynamic Routing Between Capsules
Sabour, S., Frosst, N., & Hinton, G. E. (2017) · arXiv · doi:10.48550/arxiv.1710.09829 · Cited by 4,167 · cited 1×
“An interesting work using capsule layers (a concept recently introduced by Sabour et al. [76]) is TextCaps [77], a CNN with 3 convolutional layers and two capsule layers, which obtain accuracies of 95.36% in EMNIST Letters and 99.79% in EMNIST Digits.”
MethodInfluentialMulti-column deep neural networks for image classification
Ciresan, D., Meier, U., & Schmidhuber, J. (2012) · 2012 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2012.6248110 · Cited by 3,720
Best practices for convolutional neural networks applied to visual document analysis
Simard, P. Y., Steinkraus, D., & Platt, J. C. (2005) · Seventh International Conference on Document Analysis and Recognition, 2003. Proceedings. · doi:10.1109/icdar.2003.1227801 · Cited by 2,741 · cited 2×
“87% NN 2-layer 800 hidden units, MSE loss [9] 0.”
BackgroundSalakhutdinov, R., & Hinton, G. E. (2009) · International Conference on Artificial Intelligence and Statistics · Cited by 2,605
Lee, H., Grosse, R., Ranganath, R., & Ng, A. Y. (2009) · Proceedings of the 26th Annual International Conference on Machine Learning · doi:10.1145/1553374.1553453 · Cited by 2,412 · cited 2×
“Lee et al. [54] described convolutional deep belief networks with probabilistic max-pooling achieving an error rate of 0.82%.”
BackgroundInfluentialWhat is the best multi-stage architecture for object recognition?
Jarrett, K., Kavukcuoglu, K., Ranzato, M. A., & LeCun, Y. (2009) · 2009 IEEE 12th International Conference on Computer Vision · doi:10.1109/iccv.2009.5459469 · Cited by 2,153 · cited 3×
“…[39] 0.47% CNN (3 conv, 1 dense), stochastic pooling [25] 0.49% CNN (2 conv, 1 dense, relu) with dropout [24] 0.52% CNN, unsup pretraining [17] 0.53% CNN (2 conv, 1 dense, relu) with DropConnect [24] 0.57% SparseNet + SVM [15] 0.59% CNN (2 conv, 1 dense), unsup pretraining [16] 0.60%…”
MethodInfluential(2013) · Cited by 2,056
Research works printed but not linked: 5
- 23. Kasun, L.L.C.; Zhou, H.; Huang, G.B.; Vong, C.M. Representational learning with extreme learning machine for big data. IEEE Intell. Syst. 2013 , 28 , 31-34.
- 39. Lin, M.; Chen, Q.; Yan, S. Network In Network. In Proceedings of the 2nd International Conference on Learning Representations, Banff, AB, Canada, 14-16 April 2014.
- 42. Calderón, A.; Roa-Valle, S.; Victorino, J. Handwritten digit recognition using convolutional neural networks and Gabor filters. In Proceedings of the 2003 International Conference on Computational Intelligence, Cancun, Mexico, 19-21 May 2003.
- 73. Peng, Y.; Yin, H. Markov Random Field Based Convolutional Neuralx Networks for Image Classification. In IDEAL 2017: Intelligent Data Engineering and Automated Learning ; Lecture Notes in Computer Science; Yin, H., Gao, Y., Chen, S., Wen, Y., Cai, G., Gu, T., Du, J., Tallón-Ballesteros, A., Zhang, M., Eds.; Springer: Guilin, China, 2017; Volume 10585, pp. 387-396.
- 75. Mor, S.S.; Solanki, S.; Gupta, S.; Dhingra, S.; Jain, M.; Saxena, R. Handwritten text recognition: With deep learning and Android. Int. J. Eng. Adv. Technol. 2019 , 8 , 172-178.
Cited by
Items 201–300 of 307
Handwritten Word Recognition Using Deep Learning Methods
Lagios, V., Perikos, I., & Hatzilygeroudis, I. (2023) · IFIP advances in information and communication technology · doi:10.1007/978-3-031-34171-7_28 · Cited by 1
Xiao, Y. P. (2023) · International Journal of Systems Assurance Engineering and Management · doi:10.1007/s13198-023-02015-x · Cited by 1
Woo, M., Zulkernine, F. H., & Abdulsalam, H. M. (2023) · IEEE Annual International Computer Software and Applications Conference (COMPSAC) · doi:10.1109/compsac57700.2023.00046 · Cited by 1
Use of Machine Learning Algorithms to Analyze the Digit Recognizer Problem in an Effective Manner
Shakoor, U., Mim, S. S., & Logofătu, D. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-44201-8_40 · Cited by 1
Recognition of Handwritten Character using Recognition Model based on SVM
Punia, A. (2023) · International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud) (I-SMAC) · doi:10.1109/i-smac58438.2023.10290409 · Cited by 1
Kirupakaran, A. M., Yadav, K. S., Saidulu, N., Barlaskar, S. A., & Laskar, R. H. (2024) · Multimedia Tools and Applications · doi:10.1007/s11042-024-18261-5 · Cited by 1
“In the case of handwritten characters, the standard handwritten databases – EMNIST and NIST is used [4, 21].”
Rao, I. S., Shirgire, P., Sanganwar, S., Vyawhare, K., & Vispute, S. R. (2024) · Lecture notes in electrical engineering · doi:10.1007/978-981-99-7954-7_31 · Cited by 1
Handwritten Digit Classification based on Ensemble Majority Voting Classifier
Alsharif, M., Abdo, A. A., & Bozed, K. A. (2023) · International Conference on Systems and Control (ICSC) · doi:10.1109/icsc58660.2023.10449872 · Cited by 1
“In this particular study, the MINST dataset was chosen, which consists of 70,000 handwritten figures created by 750 Census Bureau employees and high school students [13].”
MethodAn Analysis of Federated Learning Performance in Mobile Networks
Bochie, K., Sammarco, M., & Campista, M. E. M. (2023) · IEEE Virtual Conference on Communications (VCC) · doi:10.1109/vcc60689.2023.10474642 · Cited by 1
Comparative analysis of CNN models for handwritten digit recognition
Banaszewska, K., & Plechawska–Wójcik, M. (2024) · Journal of Computer Sciences Institute · doi:10.35784/jcsi.6239 · Cited by 1
“Każdy obraz, znormalizowany do rozmiaru 28x28 pikseli, przedstawia pojedynczą cyfrę od 0 do 9 w skali szarości (Rysunek 1) [6, 7].”
BackgroundIA&AI: Interference Analysis in Multi-core Embedded AI Systems
Oliveira, A., Moreira, G., Costa, D., Pinto, S., & Gomes, T. (2024) · Communications in computer and information science · doi:10.1007/978-981-97-9793-6_13 · Cited by 6
The Effect of Changing Image Contrast on Object Recognition by a Convolutional Neural Network
Tumakov, D. N., Tuliabaeva, D., & Elshin, L. A. (2024) · Lecture notes in networks and systems · doi:10.1007/978-981-97-7710-5_47 · Cited by 1
An Ensemble Transfer Learning Model for the Automatic Handwriting Recognition of Kurdish Letters
Qadir, A. M., Abdalla, P. A., Ghareb, M. I., Abd, D., & HamaKarim, K. M. (2025) · Iraqi Journal for Electrical And Electronic Engineering · doi:10.37917/ijeee.21.2.6 · Cited by 1
A Hybrid Approach for Denoising and Recognition of Handwritten Characters Using Deblur GAN-CNN
Bharati, P. V. (2024) · IEEE International Conference on ICT in Business Industry & Government (ICTBIG) · doi:10.1109/ictbig64922.2024.10911649 · Cited by 1
Dar, E., Rana, K., Yousaf, H. F., Asif, T. C., Hussain, M. Z., Ijaz, H., & Hasan, M. Z. (2024) · International Conference on Advances in Computation, Communication and Information Technology (ICAICCIT) · doi:10.1109/icaiccit64383.2024.10912098 · Cited by 1
“However, datasets such as EMNIST provide a much wider potential for the development of CNNs, describing the continuous elaboration of benchmarks for the recognition of handwritten characters [23][24][25].”
Enhanced OCR Techniques for Recognizing Mathematical Expressions in Scanned Documents
Al-Askary, Y. B., & Al-Momen, S. (2025) · Ibn AL- Haitham Journal For Pure and Applied Science · doi:10.30526/38.4.3640 · Cited by 1
“They can be handwritten or printed in scientific and mathematical formats (6,7).”
Hybridization of TrellisNet with CNN
Jaiswal, A., Duvvada, P., & Nair, L. S. (2021) · Lecture notes on data engineering and communications technologies · doi:10.1007/978-981-15-9647-6_59 · Cited by 1
Privacy-Preserving Serverless Edge Learning with Decentralized Small DataPreprint
Lin, S., & Lin, C.-H. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2111.14955 · Cited by 1
“We implement the prototype network in a serverless edge learning framework and compare the results with two competitors [14] to show the potential of few-shot learning algorithms in distributed deep learning training environment.”
MethodResultInfluentialTraining of Offline Handwritten Text Recognisers Using Computer-Generated TextPreprint
Ehiorobo, E. E., Koleoso, R. A., & Uwadia, C. O. (2022) · SSRN Electronic Journal · doi:10.2139/ssrn.4064782 · Cited by 1
(2021) · PeerJ Computer Science · doi:10.7287/peerj-cs.565v0.2/reviews/1 · Cited by 1
Khorramfar, M. (2023) · doi:10.36939/ir.202308301626 · Cited by 3
A New Unsupervised/Reinforcement Learning Method In Spiking Pattern Classification NetworksPreprint
Nazari, S. (2023) · Research Square · doi:10.21203/rs.3.rs-3560563/v1 · Cited by 2
Handwriting Rehabilitation Using a Haptic Joystick
Quaresma, V. A. G. (2021) · Portuguese National Funding Agency for Science, Research and Technology (RCAAP Project by FCT) · Cited by 1
Data-driven modelling of dynamic movements, applied to human gestures on a humanoid robot torso.
Liendo, F. (2024) · theses.fr (ABES) · doi:10.70675/f9832222z3927z4e6cz925ezaba04e330c40 · Cited by 2
Digit Recognition Based on Specialization, Decomposition and Holistic Processing
Joseph, M. D. G., & Elleithy, K. M. (2020) · Machine Learning and Knowledge Extraction · doi:10.3390/make2030015
“The error rates reported ranged from 0.95% to 0.24% [26].”
BackgroundResultInfluentialTiwari, R., Sen, A., Banerjee, A., & Dey, K. (2020) · International Conference on Innovations in Information Technology · doi:10.1109/iit50501.2020.9299031
“Although many deep learning-based classifications algorithms are studied in the literature for handwritten image recognition [2], most of the works are concentrated on a few widely used benchmark datasets [3].”
BackgroundQuantifying Uncertainty from Different Sources in Deep Neural Networks for Image ClassificationPreprint
Khoshsirat, A. (2020) · arXiv (Cornell University)
Khoshsirat, A. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2011.08712
“The one with the least model uncertainty, categorical cross-entropy function, is commonly used for classification of MNIST [16].”
MethodInfluentialMrs Performance evaluation of Convolution Neural network for handwritten Digit Recognition
Manjula, T. R. (2021) · SPAST Abstracts
Hardware Implementation of Spiking Neural Networks
Syutkin, A. (2021) · Spectrum Research Repository (Concordia University)
A Novel Approach for Handwritten Digit Recognition Using Multilayer Perceptron Neural Network
Datsi, T., Aznag, K., & Oirrak, A. E. (2022) · Advances in intelligent systems and computing · doi:10.1007/978-3-030-90639-9_19
Применение технологии CUDA для обучения нейронной сети Кохонена
Латыпова, Д. С., & Tumakov, D. N. (2022) · Международный журнал Программные продукты и системы · doi:10.15827/0236-235x.139.362-373
Clustering of handwritten digits by Kohonen neural network
Латыпова, Д. С., & Tumakov, D. N. (2022) · Program systems theory and applications · doi:10.25209/2079-3316-2022-13-3-241-254
Privacy preserving and performance improvement in edgecomputing using Machine learning
Bhadauria, S., Kumar, M., & Singh, N. P. (2022) · IEEE Global Conference on Computing, Power and Communication Technologies (GlobConPT) · doi:10.1109/globconpt57482.2022.9938248
“[4] launched the MNIST (Mixed National Institute of Standards and Technology) dataset, it has since been commonly used as a test bed for numerous computers proposals for learning and pattern recognition.”
Method- Retracted
A Novel Japanese Character Recognition Algorithm for Automatic Smart Courseware Generation Tasks
Feng, T. (2023) · International Conference on Communication and Electronics Systems (ICCES) · doi:10.1109/icces57224.2023.10192834
“…recent years, a large number of human-designed tools and highly efficient programs have played an important role in enabling computers to make great breakthroughs in automatic recognition and intelligent learning, especially in text pattern recognition [1, 2, 3] and character recognition [4, 5, 6].”
BackgroundMachine Vision for Automatic Recognition of Aircraft Engine Blade Coding
Li, C., Zhang, C., & Xu, W. (2023) · International Seminar on Artificial Intelligence, Networking and Information Technology (AINIT) · doi:10.1109/ainit59027.2023.10212638
Design and Development of robust OCR system for hand filled form in Office AutomationPreprint
Agrawal, A., Roy, P. S., Rakesh, V. K., Samanta, T., & Pal, S. (2024) · Research Square · doi:10.21203/rs.3.rs-3815243/v1
Learning Chemical Compositions through Music
Mallan, V. S., Alex, R., & James, A. P. (2024) · International Conference on Consumer Electronics · doi:10.1109/icce59016.2024.10444415
“For the identification of the alphabet, a deep learning model is created that uses the EMNIST (Extended Modified National Institute of Standards and Technology) dataset [10] [11] for training, testing and validation.”
MethodEvrişimli Sinir Ağı Kullanarak El yazısı Rakamların Tanımasında Hiper Parametre Analizi
Yiğit, T., ATMACA, Ş., Gürfidan, R., & Çolak, R. (2023) · Gazi Journal of Engineering Sciences · doi:10.30855/gmbd.0705s25
Local Fourier features for handwriting digit images classification
Djimeli-Tsajio, A., Noulamo, T., Jean-Pierre, L. T., Tchiotsop, D., & Nagabhushan, P. N. (2024) · International Journal of Power Electronics and Drive Systems/International Journal of Electrical and Computer Engineering · doi:10.11591/ijece.v14i3.pp2592-2601
“Although works using convolutional neural networks (CNNs) are becoming the state-of-the-art techniques for the classification of hand writing digits with interesting accuracy [3], CNNs are slower than classical classifiers, the training process is CPU intensive and time consuming.”
BackgroundInfluentialAirMouse and AirWrite Intuitive Digital Interaction
Varma, M. K. S., Chaitanya, M. U., Bhushan, M. S., Sivasai, M. J. N. D., & Viswa, K. C. (2024) · International Journal of Advanced Research in Science Communication and Technology · doi:10.48175/ijarsct-17054
Lima, G. E., Ladeira, R. D. L. R., & Penharbel, É. A. (2024) · Anais do Computer on the Beach · doi:10.14210/cotb.v15.p158-164
“Delimitando o escopo do trabalho, essa pesquisa estuda o ques-tionamento supracitado na classificação de dígitos manuscritos; escolhida pela sua aplicação em soluções de Reconhecimento Óp-tico de Caracteres e pela existência de pesquisas relacionadas para referência [10, 11].”
BackgroundA paper-based cheat-resistant multiple-choice question system with automated grading
Jean-Pierre, L. T., Djimeli-Tsajio, A., Noulamo, T., & Talla, B. F. (2024) · International Journal of Evaluation and Research in Education (IJERE) · doi:10.11591/ijere.v13i4.28324
“Although works using CNNs are becoming state-of-the-art techniques for the classification of handwriting digits with interesting accuracy [4], CNNs are slower than classical classifiers, the training process is CPU-intensive and time-consuming.”
BackgroundMethodInfluentialLiew, H. H., Ng, W. S., & Chen, H. V. (2024) · ITM Web of Conferences · doi:10.1051/itmconf/20246701031
Балалаєва, О., Чичкарьов, Є. А., Zinchenko, O. V., Сергієнко, А. В., & Kovalov, O. (2023) · Reporter of the Priazovskyi State Technical University Section Technical sciences · doi:10.31498/2225-6733.47.2023.299989
Object Counting from Images Using Deep Learning Technique
Bali, A. K., & Kumar, A. (2024) · Lecture notes in networks and systems · doi:10.1007/978-981-97-4152-6_17 · Cited by 1
Using FasterNet-HCCR to Speed up Handwritten Chinese Character Recognition
Ren, Z., Li, Y., & Yang, X. (2024) · International Conference on Intelligent Computing and Signal Processing (ICSP) · doi:10.1109/icsp62122.2024.10743265
The predictability of the effectiveness of chains of classifiers in the out-of-domain detection
Cofta, P. (2024) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2024.109682 · Cited by 1
Word Image Representation at Local and Global Levels Based on Vision Transformers
Harrath, B. E., Mhiri, M., & Cheriet, M. (2024) · International Conference on Computer Science and Engineering (UBMK) · doi:10.1109/ubmk63289.2024.10773601
“Extensive research has focused on improving handwritten text recognition accuracy through various methods, as discussed in numerous publications [11], [12].”
Handwritten Digit Recognition on Exam Papers Using Mnist and Emnist Dataset Trained Cnn ModelsPreprint
Senturk, A. (2024) · SSRN Electronic Journal · doi:10.2139/ssrn.5061151
A Learning Neural Network Implementing the K-Nearest Neighbors AlgorithmPreprint
Goltsev, A., & Holtsev, O. O. (2024) · SSRN Electronic Journal · doi:10.2139/ssrn.5072580
FedEC: Enhancing model federated averaging via two-sided method
Minh, D. T., Tran, T.-H., Le, T., & Tran, T. K. (2024) · Journal of Military Science and Technology · doi:10.54939/1859-1043.j.mst.csce8.2024.65-75
Learning a Sequence of Cursive-Style Japanese Characters in Classical Literary Works
Fujita, S., & Oyama, K. (2024) · Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC) · doi:10.1109/apsipaasc63619.2025.10849065
“The recognition of hand writing numbers in the MNIST dataset is a common learning target for neural network[6].”
Exploration of hyperparameter tuning in handwritten digit recognition datasets using CNN
Kundu, R., Sinha, A., Kumar, B., Gautam, R., Raza, M. S., & Hussain, S. A. (2025) · F1000Research · doi:10.12688/f1000research.161053.1
Comparative Study of Melanoma Disease Classification Using Deep Learning
Panchumarthi, L. Y., Gogineni, A. K., Saleti, S., Parchuri, L., & Kakarala, H. (2024) · OITS International Conference on Information Technology (OCIT) · doi:10.1109/ocit65031.2024.00077
“Baldominos et al. [21] carried out a survey on handwritten character recognition using MNIST and EMNIST, and they summarized the finest cutting-edge findings published on the MNIST dataset.”
DResNet: A Deep Residual Network for Enhancing Handwritten Character Recognition
Faldu, P., Gajjar, D., Kothadiya, D., & Rahevar, M. L. (2025) · Algorithms for intelligent systems · doi:10.1007/978-981-97-8868-2_23
LikhAI: Handwritten Text Recognition and Processing Application
Tiwari, V., Navlani, R., & Patel, M. (2025) · International Journal of Engineering Development and Research · doi:10.56975/ijedr.v13i2.301403
OPTIMIZED OCR APPROACHES FOR ACCURATE TEXT EXTRACTION IN LEGAL AND FINANCIAL DOCUMENT AUTOMATION
Nagalaxmi, B., Sujatha, P., Sandeep, G., Varun, G., & Teja, B. M. S. (2025) · Journal of Science Engineering Technology and Management Sciences · doi:10.63590/jsetms.2025.v02.i07.pp486-495 · Cited by 1
A Perceptron-Like Neural Network Implementing a Learning-Capable K-Nearest Neighbor Classifier
Goltsev, A., & Holtsev, O. O. (2025) · Lecture notes in computer science · doi:10.1007/978-3-032-02725-2_21
Handwriting Digit Recognition Using CNN
Yadav, N., Sharma, P., Singh, A., Deshmukh, A., Thakur, A., & Gora, A. (2025) · Lecture notes in networks and systems · doi:10.1007/978-3-032-06671-8_34
Conditional Machine Unlearning: Balancing Privacy and Regulation
Zou, Y., Qi, S., Zhang, Y., Yuan, Y., Gu, H., Li, P., & Yu, D. (2025) · Chinese Journal of Electronics · doi:10.23919/cje.2024.00.343
A Targeted Machine Unlearning Method for Sensitive Data in Military Helicopter Models
Kwon, H., & Baek, J.-W. (2025) · IEEE Access · doi:10.1109/access.2025.3631684
“Various unlearning methodologies have been proposed, although most studies primarily validate machine unlearning methods on benchmark datasets, such as CIFAR10 [9], SVHN [10], and MNIST [11], mainly at the level of toy experiments.”
Zhang, M., Guo, P., & Bai, C. G. (2025) · The Journal of Supercomputing · doi:10.1007/s11227-025-08102-z
Min, K., Jang, H., & Kim, S. J. (2025) · Journal of Alloys and Compounds · doi:10.1016/j.jallcom.2025.185586
Membership Inference Attacks on Hyperdimensional Computing: A Comprehensive Architectural Analysis
Karn, R. R., Knechtel, J., & Sinanoglu, O. (2025) · International Conference on Machine Learning and Cybernetics · doi:10.1109/icmlc66258.2025.11280147
Transition of AI Models in dependence of noisePreprint
Seidler, T., & Abel, M. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2506.16715
Culea, H. (. (2021) · Research Repository (Delft University of Technology) · Cited by 1
Clusiau, S. (2024) · eScholarship@McGill (McGill) · doi:10.82308/3150
Modélisation de la structure et dynamique corticale avec application en apprentissage machine
Brodeur, S. (2020) · Knowledge UdeS (Institutional Deposit of the University of Sherbrooke)
Rajatun laskentatehon neuroverkot OSINT-kuvien käsittelijänä
Koivikko, V. (2024) · Doria (University of Helsinki)
LLM-Driven Code Generation for Neural Networks on FPGAs: Bridging Python and HLS
Karn, R. R., Knechtel, J., Karri, R., & Sinanoglu, O. (2025) · IEEE International Conference on Computer Design: VLSI in Computers and Processors (ICCD) · doi:10.1109/iccd65941.2025.00090
“We use four datasets: Iris [10], Activities [11], UNSW [12], and MNIST [13].”
Saccade-Mimicking Grid Cell Network for Image Recognition
Matsuda, Y., Niwa, K., Aoki, T., Takadama, K., & Sato, H. (2026) · Journal of Advanced Computational Intelligence and Intelligent Informatics · doi:10.20965/jaciii.2026.p0078
Recognition of Handwritten Numbers Based on PVDF Array Tactile Sensors
Huai, X., Li, Y., Dai, X., Zhou, X. L., Zhao, T., & Gu, D. (2025) · IEEE International Conference on Data Science and Computer Application (ICDSCA) · doi:10.1109/icdsca67083.2025.11350045
“In existing literature, most handwritten datasets are sourced from the MNIST dataset [13] , which consists of individual images.”
Handwritten Digit Recognition on the MNIST Dataset using Convolutional Neural Networks
Yadav, A., Shukla, A., & Siddiqui, K. M. (2025) · i-manager's Journal on Data Science & Big Data Analytics (JDS). · doi:10.26634/jds.3.2.22576
Cortés, E. S. (2023) · Communities in ADDI (University of the Basque Country)
SHEN, M. (2026) · International Conference on Physics, Photonics, and Optical Engineering (ICPPOE) · doi:10.1117/12.3108462
“According to Baldominos et al. [8], the MNIST database is used to train models.”
Achaliya, P., Singh, S., & Monga, S. (2026) · SSRN Electronic Journal · doi:10.2139/ssrn.6462432
Analyse de données et modélisation méso-échelle des nuages de CO2 dans les nuits polaires martiennes
Caillé, V. (2023) · theses.fr (ABES) · doi:10.70675/c176a3c3z8323z460bz9c52z015db05cddbd
Modélisation statistique des cartes des dissimilarités locales d'images et applications
Diaw, M. (2022) · theses.fr (ABES) · doi:10.70675/c4297405zf776z4cdbzacebz7675bb545a67
MItra, R., Madhu, G. C., Manju, V. N., Singh, U. K., Brída, P., & Bhatia, V. (2026) · IEEE International Conference on Interdisciplinary Approaches in Technology and Management for Social Innovation (IATMSI) · doi:10.1109/iatmsi68868.2026.11466106
“Therefore, motivated by interest in CNNs for the 62 class EMNIST dataset [2], in this paper, we consider vanilla shallow CNNs, and enhance them using hyperparemeter-free RFFs.”
InfluentialMethodology and tooling for energy-efficient neural networks computation and optimization
Waltsburger, H. (2024) · theses.fr (ABES) · doi:10.70675/6aadb508z47bfz4dc6z95b6zbd5a77514a0a
Carboni, L. (2023) · theses.fr (ABES) · doi:10.70675/04aa8f1fz5ebdz4fe9zb3caz7c1377605d8f
An Optical Neural Network Model Based on Nonlinear Optical Effects
Debeljak, M. (2026) · Repository of the University of Ljubljana (University of Ljubljana)
CharNet-Hybrid: An Improved CNN-SVM Approach for Recognizing Handwritten Characters
Fethi, M. R., Farhaoui, O., Alaoui, A. O., Zeroual, I., & Allaoui, A. E. (2026) · Advances in Science, Technology & Innovation/Advances in science, technology & innovation · doi:10.1007/978-3-031-97825-8_77
Handwritten Digit Recognition Using Machine Learning
Hembram, J., Narayan, U., Adak, A., Perween, S., Pal, D., Guria, M., & Dey, D. (2025) · International Conference on Emerging Trends in Electronic Devices and Computational Techniques (EDCT) · doi:10.1109/edct65302.2025.11495795
“In 1998, the introduction of the MNIST dataset by Yann LeCun offered a customary yardstick and thus gave a boost towards research in the area [3].”
Dang, H.-A., Dao-Van, D., Vuong-Duc, M., & Cao-Thanh, T. (2026) · Lecture notes in networks and systems · doi:10.1007/978-3-032-07721-9_30
Balram Yadav, J. B. G. (2026) · International Journal of Advanced Research in Science Communication and Technology · doi:10.48175/ijarsct-35887
Atia, A. G., Karawia, A. A., Fouda, Y. M., & El-Gayar, M. M. (2026) · Applied Computational Intelligence and Soft Computing · doi:10.1155/acis/1125376
Hardware Usage Improvement for Small Data Problem Solving by Deep Learning Methods
Krak, I. V., Barmak, O., Kuznetsov, V., Kondratiuk, S., Стеля, О., & Kasianiuk, V. S. (2023) · International Conference on Information Control Systems & Technologies (ICST) · doi:10.55056/ceur-ws.org/vol-3513/paper10.pdf
“We tested various models including normal networks, convnets, and autoencoders on tiny datasets (fashion M.N.I.S.T. [23] and M.N.I.S.T. numbers [24]) but found that the graphics cards were not useful in this scenario.”
MethodStrategic Decoy Deployment for Mitigating Poisoning Attacks in Federated Learning
Masse, G. C. T., Benslimane, A., Tchendji, V. K., & Hemida, A. H. A. (2026) · International Conference on Communications · doi:10.1109/icc59461.2026.11588144
“The data distribution follows a non-IID partition of the MNIST dataset [13] using a Dirichlet distribution with α = 0 .”
Do Flat Representation Manifolds Lead to Improved Accuracy?
Burr, J., & Sarishvili, A. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-32335-4_14
Gunawardena, T., Rathnasinghe, E., Abesinghe, C., Mihara, B., Abeywardhana, L., Weerasinghe, L., & Selvaratnam, N. (2026) · International Conference on Electronics, Computers and Artificial Intelligence (ECAI) · doi:10.1109/ecai69016.2026.11613754
“Nevertheless, the achieved accuracy is comparable to previously reported CNN-based results on the EMNIST dataset [20].”
Munsarif, M., & Ampah-Mensah, R. (2026) · Cureus Journal of Computer Science · doi:10.7759/s44389-026-00193-w
“This framing is more appropriate for a mature benchmark such as MNIST, where peak accuracy alone is no longer a strong novelty claim [5].”
Suo, X., & Reinhold, H. (2026) · IEEE International Symposium on Parallel & Distributed Processing, Workshops and Phd Forum · doi:10.1109/ipdpsw71298.2026.00124
“The controlled experiments (Table I, Steps 3–7) used two compact deep-learning workloads: (1) a handwritten-character classification task based on the Kaggle Handwriting Recognition dataset [7], [8], and (2) a dog–cat binary image classifier [14].”
Fethi, M. R., Farhaoui, O., Alaoui, A. O., Zeroual, I., & Allaoui, A. E. (2026) · Lecture notes in networks and systems · doi:10.1007/978-3-032-28097-8_11
Evaluace metod pro lokalizaci těla řidiče ve vozidle
Chytil, F. (2026) · DSpace VŠB-TUO (VŠB-TUO)
Metadata
- Authors
- Baldominos, A., Saez, Y., & Isasi, P.
- Affiliations
- Universidad Carlos III de Madrid (ES)
- Computer Science Department, Universidad Carlos III of Madrid, 28911 Leganés, Madrid, Spain
- Corresponding author
- Baldominos, Alejandro
- Venue
- Applied Sciences
- Publisher
- MDPI AG
- Published
- 2019-08-04
- Volume
- 9
- Issue
- 15
- Pages
- 3169
- ISSN
- 2076-3417 (ISSN-L)
- OpenAlex
- W2966275995
- Cited by
- 279
- References
- 76
- Citations by year
- 2026: 20 · 2025: 39 · 2024: 64 · 2023: 54 · 2022: 41 · 2021: 36 · 2020: 24 · 2019: 1
- FWCI
- 11.41
- Type
- journal article
- Language
- en
- Open access
- gold
- dx.doi.org
- Licence
- Locations
- Applied Sciences (publishedVersion, cc-by)
- e-Archivo (Carlos III University of Madrid) (submittedVersion, cc-by)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion)
- Funders
- Ministerio de Educación, Cultura y Deporte (FPU13/03917)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Handwritten Text Recognition Techniques
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Vehicle License Plate Recognition
- Keywords
- handwritten character recognition, handwritten digit recognition, MNIST, EMNIST, convolutional neural networks, data augmentation, letter recognition
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.


