{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,12]],"date-time":"2026-06-12T00:58:34Z","timestamp":1781225914341,"version":"3.54.1"},"reference-count":47,"publisher":"Elsevier BV","license":[{"start":{"date-parts":[[2026,6,1]],"date-time":"2026-06-01T00:00:00Z","timestamp":1780272000000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.elsevier.com\/tdm\/userlicense\/1.0\/"},{"start":{"date-parts":[[2026,6,1]],"date-time":"2026-06-01T00:00:00Z","timestamp":1780272000000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.elsevier.com\/legal\/tdmrep-license"},{"start":{"date-parts":[[2026,6,1]],"date-time":"2026-06-01T00:00:00Z","timestamp":1780272000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-017"},{"start":{"date-parts":[[2026,6,1]],"date-time":"2026-06-01T00:00:00Z","timestamp":1780272000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"},{"start":{"date-parts":[[2026,6,1]],"date-time":"2026-06-01T00:00:00Z","timestamp":1780272000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-012"},{"start":{"date-parts":[[2026,6,1]],"date-time":"2026-06-01T00:00:00Z","timestamp":1780272000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,6,1]],"date-time":"2026-06-01T00:00:00Z","timestamp":1780272000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-004"}],"funder":[{"DOI":"10.13039\/501100022101","name":"University of Toronto Mississauga","doi-asserted-by":"publisher","award":["10G05"],"award-info":[{"award-number":["10G05"]}],"id":[{"id":"10.13039\/501100022101","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100022101","name":"University of Toronto Mississauga","doi-asserted-by":"publisher","award":["10G09"],"award-info":[{"award-number":["10G09"]}],"id":[{"id":"10.13039\/501100022101","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["elsevier.com","sciencedirect.com"],"crossmark-restriction":true},"short-container-title":["Knowledge-Based Systems"],"published-print":{"date-parts":[[2026,6]]},"DOI":"10.1016\/j.knosys.2026.116018","type":"journal-article","created":{"date-parts":[[2026,4,15]],"date-time":"2026-04-15T07:23:18Z","timestamp":1776237798000},"page":"116018","update-policy":"https:\/\/doi.org\/10.1016\/elsevier_cm_policy","source":"Crossref","is-referenced-by-count":1,"special_numbering":"C","title":["ReLU-DCT and LASP: Real-valued frequency-domain activations and adaptive pooling for DCT-Based CNNs"],"prefix":"10.1016","volume":"343","author":[{"ORCID":"https:\/\/orcid.org\/0009-0002-2418-9277","authenticated-orcid":false,"given":"Ibrahim Yousef","family":"Alshareef","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0754-5199","authenticated-orcid":false,"given":"Ab Al-Hadi Ab","family":"Rahman","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3888-226X","authenticated-orcid":false,"given":"Ahmad Zuri","family":"Shaameri","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1874-895X","authenticated-orcid":false,"given":"Nuzhat","family":"Khan","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mohd Shahrizal","family":"Rusli","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6238-8915","authenticated-orcid":false,"given":"Mohamed Khalafalla","family":"Hassan","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6312-2262","authenticated-orcid":false,"given":"Ali","family":"Manzak","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Muhammad Nadzir","family":"Marsono","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"78","reference":[{"key":"10.1016\/j.knosys.2026.116018_b1","doi-asserted-by":"crossref","DOI":"10.1016\/j.asoc.2021.107851","article-title":"Image classification in frequency domain with 2SReLU: A second harmonics superposition activation function","volume":"112","author":"Watanabe","year":"2021","journal-title":"Appl. Soft Comput."},{"key":"10.1016\/j.knosys.2026.116018_b2","doi-asserted-by":"crossref","first-page":"254","DOI":"10.1016\/j.neucom.2019.06.080","article-title":"A Fourier domain acceleration framework for convolutional neural networks","volume":"364","author":"Lin","year":"2019","journal-title":"Neurocomputing"},{"key":"10.1016\/j.knosys.2026.116018_b3","doi-asserted-by":"crossref","first-page":"53785","DOI":"10.1007\/s11042-023-17546-5","article-title":"A comparative analysis of various activation functions and optimizers in a convolutional neural network for hyperspectral image classification","volume":"83","author":"Seyrek","year":"2024","journal-title":"Multimedia Tools Appl."},{"issue":"2","key":"10.1016\/j.knosys.2026.116018_b4","doi-asserted-by":"crossref","first-page":"176","DOI":"10.11113\/elektrika.v24n2.651","article-title":"Optimizing spectral activation functions for spectral convolutional neural networks: balancing efficiency and fidelity","volume":"24","author":"Alshareef","year":"2025","journal-title":"ELEKTRIKA - J. Electr. Eng."},{"issue":"2","key":"10.1016\/j.knosys.2026.116018_b5","doi-asserted-by":"crossref","first-page":"140","DOI":"10.1109\/TETCI.2018.2872600","article-title":"Complex-valued neural networks with nonparametric activation functions","volume":"4","author":"Scardapane","year":"2020","journal-title":"IEEE Trans. Emerg. Top. Comput. Intell."},{"key":"10.1016\/j.knosys.2026.116018_b6","doi-asserted-by":"crossref","first-page":"879","DOI":"10.1007\/s00521-019-04296-5","article-title":"Interpretation of intelligence in CNN-pooling processes: a methodological survey","volume":"32","author":"Akhtar","year":"2020","journal-title":"Neural Comput. Appl."},{"key":"10.1016\/j.knosys.2026.116018_b7","doi-asserted-by":"crossref","DOI":"10.1155\/2022\/4067581","article-title":"Pooling operations in deep learning: from \u201cinvariable\u201d to \u201cvariable\u201d","volume":"2022","author":"Tao","year":"2022","journal-title":"Biomed. Res. Int."},{"key":"10.1016\/j.knosys.2026.116018_b8","article-title":"Spectral representations for convolutional neural networks","volume":"28","author":"Rippel","year":"2015","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"10.1016\/j.knosys.2026.116018_b9","doi-asserted-by":"crossref","DOI":"10.1016\/j.asoc.2026.114599","article-title":"End-to-end discrete cosine transform integration in spectral convolutional neural networks for resource-efficient deep learning","author":"Alshareef","year":"2026","journal-title":"Appl. Soft Comput."},{"key":"10.1016\/j.knosys.2026.116018_b10","series-title":"MNIST handwritten digit database","author":"LeCun","year":"2010"},{"key":"10.1016\/j.knosys.2026.116018_b11","doi-asserted-by":"crossref","DOI":"10.1007\/s00521-025-11376-2","article-title":"Lightweight and high-precision spectral convolutional neural network model for custom 94-class ASCII character recognition","author":"Alshareef","year":"2025","journal-title":"Neural Comput. Appl."},{"key":"10.1016\/j.knosys.2026.116018_b12","series-title":"Journal of Physics: Conference Series","article-title":"Sigmoid activation function in selecting the best model of artificial neural networks","volume":"vol. 1471, no. 1","author":"Pratiwi","year":"2020"},{"key":"10.1016\/j.knosys.2026.116018_b13","article-title":"The improved modified extended tanh-function method to develop the exact travelling wave solutions of a family of 3D fractional WBBM equations","volume":"41","author":"Ananna","year":"2022","journal-title":"Results Phys."},{"key":"10.1016\/j.knosys.2026.116018_b14","unstructured":"V. Nair, G.E. Hinton, Rectified linear units improve restricted Boltzmann machines, in: Proceedings of the 27th International Conference on Machine Learning, ICML-10, 2010, pp. 807\u2013814."},{"key":"10.1016\/j.knosys.2026.116018_b15","unstructured":"A.L. Maas, A.Y. Hannun, A.Y. Ng, Rectifier nonlinearities improve neural network acoustic models, in: Proceedings of the 30th International Conference on Machine Learning, vol. 30, Atlanta, GA, USA, 2013, pp. 16\u201321."},{"key":"10.1016\/j.knosys.2026.116018_b16","doi-asserted-by":"crossref","unstructured":"K. He, X. Zhang, S. Ren, J. Sun, Delving deep into rectifiers: Surpassing human-level performance on ImageNet classification, in: Proceedings of the IEEE International Conference on Computer Vision, ICCV, 2015, pp. 1026\u20131034.","DOI":"10.1109\/ICCV.2015.123"},{"key":"10.1016\/j.knosys.2026.116018_b17","unstructured":"M. Arjovsky, A. Shah, Y. Bengio, Unitary evolution recurrent neural networks, in: International Conference on Machine Learning, 2016, pp. 1120\u20131128."},{"key":"10.1016\/j.knosys.2026.116018_b18","series-title":"On complex valued convolutional neural networks","author":"Guberman","year":"2016"},{"key":"10.1016\/j.knosys.2026.116018_b19","unstructured":"C. Trabelsi, O. Bilaniuk, Y. Zhang, D. Serdyuk, S. Subramanian, J.F. Santos, S. Mehri, N. Rostamzadeh, Y. Bengio, C.J. Pal, Deep complex networks, in: International Conference on Learning Representations, ICLR, 2018."},{"key":"10.1016\/j.knosys.2026.116018_b20","doi-asserted-by":"crossref","first-page":"152","DOI":"10.1016\/j.neucom.2019.06.094","article-title":"Spectral-based convolutional neural network without multiple spatial-frequency domain switchings","volume":"364","author":"Ayat","year":"2019","journal-title":"Neurocomputing"},{"key":"10.1016\/j.knosys.2026.116018_b21","series-title":"What\u2019s wrong with convolutional nets? MIT brain and cognitive sciences - fall colloquium series","author":"Hinton","year":"2014"},{"key":"10.1016\/j.knosys.2026.116018_b22","series-title":"Stochastic pooling for regularization of deep convolutional neural networks","author":"Zeiler","year":"2013"},{"key":"10.1016\/j.knosys.2026.116018_b23","series-title":"Maxout networks","author":"Goodfellow","year":"2013"},{"key":"10.1016\/j.knosys.2026.116018_b24","series-title":"Deeply-supervised nets","author":"Lee","year":"2014"},{"key":"10.1016\/j.knosys.2026.116018_b25","series-title":"Network in network","author":"Lin","year":"2013"},{"issue":"3","key":"10.1016\/j.knosys.2026.116018_b26","doi-asserted-by":"crossref","first-page":"391","DOI":"10.1088\/0954-898X_14_3_302","article-title":"Statistics of natural image categories","volume":"14","author":"Torralba","year":"2003","journal-title":"Network"},{"key":"10.1016\/j.knosys.2026.116018_b27","doi-asserted-by":"crossref","unstructured":"K. Xu, M. Qin, F. Sun, Y. Wang, Y.K. Chen, F. Ren, Learning in the frequency domain, in: Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, 2020, pp. 1740\u20131749.","DOI":"10.1109\/CVPR42600.2020.00181"},{"key":"10.1016\/j.knosys.2026.116018_b28","doi-asserted-by":"crossref","unstructured":"M. Ehrlich, L.S. Davis, Deep residual learning in the JPEG transform domain, in: Proceedings of the IEEE\/CVF International Conference on Computer Vision, 2019, pp. 3484\u20133493.","DOI":"10.1109\/ICCV.2019.00358"},{"key":"10.1016\/j.knosys.2026.116018_b29","series-title":"ICASSP 2024\u2014IEEE International Conference on Acoustics, Speech and Signal Processing","first-page":"7455","article-title":"Adversarial robustness of convolutional models learned in the frequency domain","author":"Chaudhury","year":"2024"},{"key":"10.1016\/j.knosys.2026.116018_b30","doi-asserted-by":"crossref","unstructured":"S.M. Rizvi, A.A.H. Ab Rahman, M. Khalil-Hani, S.O. Ayat, A low-complexity complex-valued activation function for fast and accurate spectral accurate spectral domain convolutional neural network, Indones. J. Electr. Eng. Informatics (IJEEI) 9 (1) 173\u2013184.","DOI":"10.52549\/ijeei.v9i1.2737"},{"issue":"11","key":"10.1016\/j.knosys.2026.116018_b31","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":"10.1016\/j.knosys.2026.116018_b32","series-title":"Binarized neural networks: training deep neural networks with weights and activations constrained to 1 or -1","author":"Courbariaux","year":"2016"},{"key":"10.1016\/j.knosys.2026.116018_b33","first-page":"4107","article-title":"Binarized neural networks","volume":"vol. 29","author":"Hubara","year":"2016"},{"key":"10.1016\/j.knosys.2026.116018_b34","series-title":"Proc. IEEE 34th Int. Conf. Comput. Des","first-page":"678","article-title":"DSCNN: hardware-oriented optimization for stochastic computing based deep convolutional neural networks","author":"Li","year":"2016"},{"key":"10.1016\/j.knosys.2026.116018_b35","series-title":"Proc. IEEE Int. Conf. Rebooting Comput","first-page":"1","article-title":"Designing reconfigurable large-scale deep learning systems using stochastic computing","author":"Ren","year":"2016"},{"key":"10.1016\/j.knosys.2026.116018_b36","series-title":"Proc. Des. Autom. Test Eur. Conf. Exhib","first-page":"250","article-title":"Structural design optimization for deep convolutional neural networks using stochastic computing","author":"Li","year":"2017"},{"key":"10.1016\/j.knosys.2026.116018_b37","series-title":"Proc. Int. Joint Conf. Neural Netw","first-page":"1230","article-title":"Hardware-driven nonlinear activation for stochastic computing based deep convolutional neural networks","author":"Li","year":"2017"},{"issue":"2","key":"10.1016\/j.knosys.2026.116018_b38","doi-asserted-by":"crossref","first-page":"405","DOI":"10.1145\/3093315.3037746","article-title":"SC-DCNN: highly-scalable deep convolutional neural network using stochastic computing","volume":"51","author":"Ren","year":"2017","journal-title":"ACM SIGOPS Oper. Syst. Rev."},{"key":"10.1016\/j.knosys.2026.116018_b39","series-title":"Proc. 19th Int. Symp. Qual. Electron. Des","first-page":"314","article-title":"An area and energy efficient design of domain-wall memory-based deep convolutional neural networks using stochastic computing","author":"Ma","year":"2018"},{"key":"10.1016\/j.knosys.2026.116018_b40","doi-asserted-by":"crossref","first-page":"1072","DOI":"10.1016\/j.neucom.2017.09.046","article-title":"FP-BNN: binarized neural network on FPGA","volume":"275","author":"Liang","year":"2018","journal-title":"Neurocomputing"},{"key":"10.1016\/j.knosys.2026.116018_b41","series-title":"Very efficient training of convolutional neural networks using fast fourier transform and overlap-and-add","author":"Highlander","year":"2016"},{"key":"10.1016\/j.knosys.2026.116018_b42","series-title":"Proc. 54th ACM\/EDAC\/IEEE Des. Autom. Conf","first-page":"1","article-title":"Design of an energy-efficient accelerator for training of convolutional neural networks using frequency-domain computation","author":"Ko","year":"2017"},{"key":"10.1016\/j.knosys.2026.116018_b43","first-page":"786","article-title":"FCNN: Fourier convolutional neural networks","volume":"vol. 17","author":"Pratt","year":"2017"},{"key":"10.1016\/j.knosys.2026.116018_b44","series-title":"SpecNet: spectral domain convolutional neural network","author":"Guan","year":"2019"},{"issue":"4","key":"10.1016\/j.knosys.2026.116018_b45","doi-asserted-by":"crossref","first-page":"4499","DOI":"10.1007\/s10489-022-03756-1","article-title":"Computation and memory optimized spectral domain convolutional neural network for throughput and energy-efficient inference","volume":"53","author":"Rizvi","year":"2023","journal-title":"Appl. Intell."},{"issue":"12","key":"10.1016\/j.knosys.2026.116018_b46","doi-asserted-by":"crossref","first-page":"2132","DOI":"10.3390\/math10122132","article-title":"End-to-end training of deep neural networks in the Fourier domain","volume":"10","author":"F\u00fcl\u00f6p","year":"2022","journal-title":"Mathematics"},{"key":"10.1016\/j.knosys.2026.116018_b47","doi-asserted-by":"crossref","DOI":"10.1016\/j.array.2025.100594","article-title":"A real-valued DCT-based spectral CNN architecture for efficient edge deep learning","author":"Alshareef","year":"2025","journal-title":"Array"}],"container-title":["Knowledge-Based Systems"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/api.elsevier.com\/content\/article\/PII:S0950705126007446?httpAccept=text\/xml","content-type":"text\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/api.elsevier.com\/content\/article\/PII:S0950705126007446?httpAccept=text\/plain","content-type":"text\/plain","content-version":"vor","intended-application":"text-mining"}],"deposited":{"date-parts":[[2026,6,12]],"date-time":"2026-06-12T00:13:55Z","timestamp":1781223235000},"score":1,"resource":{"primary":{"URL":"https:\/\/linkinghub.elsevier.com\/retrieve\/pii\/S0950705126007446"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,6]]},"references-count":47,"alternative-id":["S0950705126007446"],"URL":"https:\/\/doi.org\/10.1016\/j.knosys.2026.116018","relation":{},"ISSN":["0950-7051"],"issn-type":[{"value":"0950-7051","type":"print"}],"subject":[],"published":{"date-parts":[[2026,6]]},"assertion":[{"value":"Elsevier","name":"publisher","label":"This article is maintained by"},{"value":"ReLU-DCT and LASP: Real-valued frequency-domain activations and adaptive pooling for DCT-Based CNNs","name":"articletitle","label":"Article Title"},{"value":"Knowledge-Based Systems","name":"journaltitle","label":"Journal Title"},{"value":"https:\/\/doi.org\/10.1016\/j.knosys.2026.116018","name":"articlelink","label":"CrossRef DOI link to publisher maintained version"},{"value":"article","name":"content_type","label":"Content Type"},{"value":"\u00a9 2026 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.","name":"copyright","label":"Copyright"}],"article-number":"116018"}}