A Deep Single-Pass Learning for Recognition of Handwritten Digits

Simple item page
dc.contributor.authorThongsuwan, Setthanun
dc.contributor.authorJaiyen, Saichon
dc.date.accessioned2026-08-06T10:36:31Z
dc.date.available2026-08-06T10:36:31Z
dc.date.issued2022-03-01
dc.description.abstractWe describe a deep learning model-Deep Single-Pass Learning (DSPL)-that can learn a data set, with a single pass for recognition, and predict with high accuracy, when evaluated for visual recognition of handwritten digits. DSPL consists of several stacked convolutional layers to learn features automatically and Extreme gradient boosting (XGBoost) was set as the last layer for predicting class labels. The learning time complexity is O(Lc<sup>2</sup> mnpq), or less than the learning time of deep learning-Convolutional Neural Networks (CNNs). The network does not need iteration to re-adjust weights during feature learning. Tests showed that our model provided better accuracy than other models i.e. CNNs, XGBoost, LR, ETC, GBC, RFC, GNB, and DTC, including MLP and SVC families: in the worst case, DSPL provided 99.95% accuracy.
dc.identifier.citationThai Journal of Mathematics, 20(1), 293-304, 2022
dc.identifier.issn16860209
dc.identifier.other2-s2.0-85130697041
dc.identifier.urihttps://dspace.kmitl.ac.th/handle/123456789/13091
dc.sourceThai Journal of Mathematics
dc.subjectconvolutional neural networks
dc.subjectdeep single-pass learning
dc.subjecthandwriting recognition
dc.subjectpattern recognition
dc.subjectxgboost
dc.titleA Deep Single-Pass Learning for Recognition of Handwritten Digits
dc.typeArticle

Files

Collections

Research