{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,16]],"date-time":"2026-07-16T16:07:41Z","timestamp":1784218061571,"version":"3.55.0"},"reference-count":40,"publisher":"Elsevier BV","license":[{"start":{"date-parts":[[2026,10,1]],"date-time":"2026-10-01T00:00:00Z","timestamp":1790812800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.elsevier.com\/tdm\/userlicense\/1.0\/"},{"start":{"date-parts":[[2026,10,1]],"date-time":"2026-10-01T00:00:00Z","timestamp":1790812800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.elsevier.com\/legal\/tdmrep-license"},{"start":{"date-parts":[[2026,10,1]],"date-time":"2026-10-01T00:00:00Z","timestamp":1790812800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-017"},{"start":{"date-parts":[[2026,10,1]],"date-time":"2026-10-01T00:00:00Z","timestamp":1790812800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"},{"start":{"date-parts":[[2026,10,1]],"date-time":"2026-10-01T00:00:00Z","timestamp":1790812800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-012"},{"start":{"date-parts":[[2026,10,1]],"date-time":"2026-10-01T00:00:00Z","timestamp":1790812800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,10,1]],"date-time":"2026-10-01T00:00:00Z","timestamp":1790812800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-004"}],"content-domain":{"domain":["elsevier.com","sciencedirect.com"],"crossmark-restriction":true},"short-container-title":["Applied Soft Computing"],"published-print":{"date-parts":[[2026,10]]},"DOI":"10.1016\/j.asoc.2026.115892","type":"journal-article","created":{"date-parts":[[2026,7,8]],"date-time":"2026-07-08T00:03:19Z","timestamp":1783468999000},"page":"115892","update-policy":"https:\/\/doi.org\/10.1016\/elsevier_cm_policy","source":"Crossref","is-referenced-by-count":0,"special_numbering":"PC","title":["Non-Newtonian mean-based logarithmic activation functions for deep learning"],"prefix":"10.1016","volume":"202","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-1589-2593","authenticated-orcid":false,"given":"Alper","family":"Ekinci","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1644-0476","authenticated-orcid":false,"given":"Abdullah","family":"Elen","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9272-5572","authenticated-orcid":false,"given":"Selcuk","family":"Bas","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"78","reference":[{"issue":"1","key":"10.1016\/j.asoc.2026.115892_bib0005","first-page":"87","article-title":"Adaptive hybrid activation function for deep neural networks","volume":"2022","author":"Bodyanskiy","year":"2022","journal-title":"Syst. Res. Inf. Technol."},{"issue":"1","key":"10.1016\/j.asoc.2026.115892_bib0010","first-page":"10","article-title":"Overview of configuring adaptive activation functions for deep neural networks: a comparative study","volume":"3","author":"Haoxiang","year":"2021","journal-title":"J. Univers. Computer Commun. Technol."},{"issue":"1","key":"10.1016\/j.asoc.2026.115892_bib0015","doi-asserted-by":"crossref","first-page":"01","DOI":"10.33140\/ACPMS.01.01.04","article-title":"Amplifying sine unit: an oscillatory activation function for DeepNeural networks to recover nonlinear oscillations efficiently","volume":"1","author":"Rahman","year":"2024","journal-title":"Ann. Comp. Phy. Material Sci."},{"issue":"3","key":"10.1016\/j.asoc.2026.115892_bib0020","doi-asserted-by":"crossref","first-page":"311","DOI":"10.47839\/ijc.22.3.3225","article-title":"Learnable extended activation function for deep neural networks","volume":"22","author":"Bodyanskiy","year":"2023","journal-title":"Int. J. Comput."},{"key":"10.1016\/j.asoc.2026.115892_bib0025","series-title":"Proceedings of the Royal Society A","article-title":"Locally adaptive activation functions with slope recovery for deep and physics-informed neural networks","author":"Jagtap","year":"2020"},{"issue":"1","key":"10.1016\/j.asoc.2026.115892_bib0030","doi-asserted-by":"crossref","first-page":"e0243915","DOI":"10.1371\/journal.pone.0243915","article-title":"On transformative adaptive activation functions in neural networks for gene expression inference","volume":"16","author":"Kunc","year":"2021","journal-title":"PLOS ONE"},{"issue":"5","key":"10.1016\/j.asoc.2026.115892_bib0035","doi-asserted-by":"crossref","first-page":"1027","DOI":"10.3390\/sym14051027","article-title":"An efficient asymmetric nonlinear activation function for deep neural networks","volume":"14","author":"Chai","year":"2022","journal-title":"Symmetry"},{"key":"10.1016\/j.asoc.2026.115892_bib0040","doi-asserted-by":"crossref","first-page":"7","DOI":"10.35784\/jcsi.3069","article-title":"Comparison of the effectiveness of artificial neural networks for various activation functions","volume":"26","author":"Florek","year":"2023","journal-title":"J. Comput. Sci. Inst."},{"issue":"3","key":"10.1016\/j.asoc.2026.115892_bib0045","doi-asserted-by":"crossref","first-page":"57","DOI":"10.9790\/0853-2003015759","article-title":"A survey on different self-gating techniques of activation functions in neural networks","volume":"20","author":"Jahan","year":"2025","journal-title":"IOSR J. Electr. Electron. Eng."},{"issue":"177","key":"10.1016\/j.asoc.2026.115892_bib0050","first-page":"13","article-title":"\u201cModified satlin activation function for deep neural networks","volume":"2","author":"Bodyanskyi","year":"2024","journal-title":"Inf. Process. Syst."},{"issue":"5","key":"10.1016\/j.asoc.2026.115892_bib0055","first-page":"4345","article-title":"Wavelets as activation functions in neural networks","volume":"42","author":"Herrera","year":"2021","journal-title":"J. Intell. Fuzzy Syst."},{"issue":"3","key":"10.1016\/j.asoc.2026.115892_bib0060","doi-asserted-by":"crossref","first-page":"246","DOI":"10.3390\/axioms12030246","article-title":"NIPUNA: a novel optimizer activation function for deep neural networks","volume":"12","author":"Madhu","year":"2023","journal-title":"Axioms"},{"issue":"7","key":"10.1016\/j.asoc.2026.115892_bib0065","doi-asserted-by":"crossref","first-page":"e0306420","DOI":"10.1371\/journal.pone.0306420","article-title":"Self-learning activation functions to increase accuracy of privacy-preserving convolutional neural networks with homomorphic encryption","volume":"19","author":"Pulido-Gaytan","year":"2024","journal-title":"PLOS ONE"},{"issue":"1","key":"10.1016\/j.asoc.2026.115892_bib0070","first-page":"93","article-title":"On approximation by neural networks with optimized activation functions and fixed weights","volume":"39","author":"Yu","year":"2023","journal-title":"Anal. Theory Appl."},{"key":"10.1016\/j.asoc.2026.115892_bib0075","article-title":"Custom nonlinear activation functions for pothole and speed bump image classification using convolutional neural networks","volume":"22","author":"Jawale","year":"2025","journal-title":"NED Univ. J. Res."},{"issue":"7","key":"10.1016\/j.asoc.2026.115892_bib0080","doi-asserted-by":"crossref","first-page":"1810","DOI":"10.1364\/OL.451287","article-title":"Programmable low-threshold optical nonlinear activation functions for photonic neural networks","volume":"47","author":"Huang","year":"2022","journal-title":"Opt. Lett."},{"issue":"2","key":"10.1016\/j.asoc.2026.115892_bib0085","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1145\/3607533","article-title":"Survey on activation functions for optical neural networks","volume":"56","author":"Destras","year":"2023","journal-title":"ACM Comput. Surv."},{"issue":"2","key":"10.1016\/j.asoc.2026.115892_bib0090","doi-asserted-by":"crossref","DOI":"10.1615\/JMachLearnModelComput.2024056966","article-title":"Neural networks with trainable matrix activation functions","volume":"6","author":"Liu","year":"2025","journal-title":"J. Mach. Learn. Model. Comput."},{"issue":"70","key":"10.1016\/j.asoc.2026.115892_bib0095","doi-asserted-by":"crossref","first-page":"95","DOI":"10.4114\/intartif.vol25iss70pp95-109","article-title":"A large-scale study of activation functions in modern deep neural network architectures for efficient convergence","volume":"25","author":"Rasamoelina","year":"2022","journal-title":"Intel. Artif."},{"key":"10.1016\/j.asoc.2026.115892_bib0100","doi-asserted-by":"crossref","first-page":"119503","DOI":"10.1016\/j.eswa.2023.119503","article-title":"Detection and classification of pneumonia using novel superior exponential activation function in convolutional neural networks","volume":"217","author":"Kilicarslan","year":"2023","journal-title":"Expert Syst. Appl."},{"issue":"24","key":"10.1016\/j.asoc.2026.115892_bib0105","doi-asserted-by":"crossref","first-page":"18451","DOI":"10.1007\/s00500-023-09279-2","article-title":"Alpha-SechSig and alpha-TanhSig: two novel non-monotonic activation functions","volume":"27","author":"Kozkurt","year":"2023","journal-title":"Soft Comput."},{"key":"10.1016\/j.asoc.2026.115892_bib0110","series-title":"Non-Newtonian Calculus","author":"Grossman","year":"1972"},{"key":"10.1016\/j.asoc.2026.115892_bib0115","first-page":"1","article-title":"A generalization on weighted means and convex functions with respect to the non-newtonian calculus","author":"Kadak","year":"2016","journal-title":"Int. J. Anal."},{"issue":"3","key":"10.1016\/j.asoc.2026.115892_bib0120","first-page":"287","article-title":"Hermite-Hadamard-Fejer inequality and some new inequalities via star-calculus","volume":"10","author":"Unluyol","year":"2019","journal-title":"Sigma J. Eng. Nat. Sci."},{"issue":"1","key":"10.1016\/j.asoc.2026.115892_bib0125","doi-asserted-by":"crossref","first-page":"36","DOI":"10.1016\/j.jmaa.2007.03.081","article-title":"Multiplicative calculus and its applications","volume":"337","author":"Bashirov","year":"2008","journal-title":"J. Math. Anal. Appl."},{"issue":"4","key":"10.1016\/j.asoc.2026.115892_bib0130","first-page":"717","article-title":"Hermite-hadamard type inequalities in the framework of non-newtonian convex functions","volume":"24","author":"Ekinci","year":"2025","journal-title":"Appl. Comput. Math."},{"key":"10.1016\/j.asoc.2026.115892_bib0135","doi-asserted-by":"crossref","first-page":"81","DOI":"10.1007\/s11082-023-05569-z","article-title":"New geometric magnetic energy according to geometric frenet formulas","volume":"56","author":"Ekinci","year":"2024","journal-title":"Opt. Quantum Electron."},{"key":"10.1016\/j.asoc.2026.115892_bib0140","series-title":"Learning Multiple Layers of Features From Tiny Images","author":"Krizhevsky","year":"2009"},{"key":"10.1016\/j.asoc.2026.115892_bib0145","author":"Xiao"},{"key":"10.1016\/j.asoc.2026.115892_bib0150","author":"Netzer"},{"key":"10.1016\/j.asoc.2026.115892_bib0160","series-title":"2019 IEEE International Conference on Industrial Engineering and Engineering Management (IEEM)","first-page":"1471","article-title":"A study of applying deep learning-based weighted combinations to improve defect prediction accuracy and effectiveness","author":"Huang","year":"2019"},{"key":"10.1016\/j.asoc.2026.115892_bib0165","author":"Goyal"},{"key":"10.1016\/j.asoc.2026.115892_bib0170","doi-asserted-by":"crossref","DOI":"10.1016\/j.cpc.2024.109428","article-title":"Simple yet effective adaptive activation functions for physics-informed neural networks","volume":"307","author":"Zhang","year":"2025","journal-title":"Comput. Phys. Commun."},{"issue":"1","key":"10.1016\/j.asoc.2026.115892_bib0175","doi-asserted-by":"crossref","first-page":"680","DOI":"10.1109\/TR.2023.3284857","article-title":"Application of weighted combinations of activation functions to defect prediction in software development","volume":"73","author":"Su","year":"2023","journal-title":"IEEE Trans. Reliab."},{"key":"10.1016\/j.asoc.2026.115892_bib0180","doi-asserted-by":"crossref","DOI":"10.1016\/j.chaos.2024.115603","article-title":"Reducing overfitting in deep learning intrusion detection for power systems with CTGAN","volume":"188","author":"Agarwal","year":"2024","journal-title":"Chaos Solit. Fractals"},{"issue":"2","key":"10.1016\/j.asoc.2026.115892_bib0185","doi-asserted-by":"crossref","first-page":"12","DOI":"10.63180\/jcsra.thestap.2025.2.2","article-title":"\u201cAnalyzing cybersecurity risks and threats in IT infrastructure based on NIST framework","volume":"2025","author":"Aljumaiah","year":"2025","journal-title":"J. Cyber Secur. Risk Audit."},{"key":"10.1016\/j.asoc.2026.115892_bib0190","series-title":"2025 International Conference on Robotics and Mechatronics (ICRM)","first-page":"1","article-title":"\u201cSpoken-intent classification using hybrid activation function","author":"Hegde","year":"2025"},{"key":"10.1016\/j.asoc.2026.115892_bib0195","series-title":"2025 6th International Conference on Mobile Computing and Sustainable Informatics (ICMCSI)","first-page":"1159","article-title":"\u201dDeep fake detection with hybrid activation function enabled adaptive milvus optimization-based deep convolutional neural network","author":"Mashetty","year":"2025"},{"key":"10.1016\/j.asoc.2026.115892_bib0200","doi-asserted-by":"crossref","first-page":"4282","DOI":"10.1038\/s41598-025-34406-4","article-title":"A hybrid CNN-transformer model with adaptive activation function for potato leaf disease classification","volume":"16","author":"Mondal","year":"2026","journal-title":"Sci. Rep."},{"issue":"4","key":"10.1016\/j.asoc.2026.115892_bib0205","doi-asserted-by":"crossref","first-page":"90","DOI":"10.3390\/jcp5040090","article-title":"AI-powered security for IoT ecosystems: a hybrid deep learning approach to anomaly detection","volume":"5","author":"Kumar","year":"2025","journal-title":"J. Cybersecur. Priv."}],"container-title":["Applied Soft Computing"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/api.elsevier.com\/content\/article\/PII:S1568494626013402?httpAccept=text\/xml","content-type":"text\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/api.elsevier.com\/content\/article\/PII:S1568494626013402?httpAccept=text\/plain","content-type":"text\/plain","content-version":"vor","intended-application":"text-mining"}],"deposited":{"date-parts":[[2026,7,16]],"date-time":"2026-07-16T15:47:40Z","timestamp":1784216860000},"score":1,"resource":{"primary":{"URL":"https:\/\/linkinghub.elsevier.com\/retrieve\/pii\/S1568494626013402"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,10]]},"references-count":40,"alternative-id":["S1568494626013402"],"URL":"https:\/\/doi.org\/10.1016\/j.asoc.2026.115892","relation":{},"ISSN":["1568-4946"],"issn-type":[{"value":"1568-4946","type":"print"}],"subject":[],"published":{"date-parts":[[2026,10]]},"assertion":[{"value":"Elsevier","name":"publisher","label":"This article is maintained by"},{"value":"Non-Newtonian mean-based logarithmic activation functions for deep learning","name":"articletitle","label":"Article Title"},{"value":"Applied Soft Computing","name":"journaltitle","label":"Journal Title"},{"value":"https:\/\/doi.org\/10.1016\/j.asoc.2026.115892","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":"115892"}}