{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,10,23]],"date-time":"2025-10-23T02:41:36Z","timestamp":1761187296565,"version":"build-2065373602"},"reference-count":32,"publisher":"MDPI AG","issue":"11","license":[{"start":{"date-parts":[[2025,10,22]],"date-time":"2025-10-22T00:00:00Z","timestamp":1761091200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Heilongjiang Provincial Basic Scientific Research Project of Higher Education Institutions in 2022","award":["1452MSYYB008"],"award-info":[{"award-number":["1452MSYYB008"]}]},{"name":"2024 Shuanggua\u2013Shuangneng Construction Project of Mudanjiang Normal University","award":["2024SGSN015"],"award-info":[{"award-number":["2024SGSN015"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Computation"],"abstract":"<jats:p>Residual Networks (ResNet) address the vanishing gradient problem through skip connections and have become a fundamental architecture for computer vision tasks. However, standard convolutional layers exhibit limited capacity in modeling complex nonlinear relationships. We present EKAResNet, a residual backbone enhanced with a spline-based Kolmogorov\u2013Arnold Network (KAN) head. Specifically, we introduce a KAN-based Feature Classification Module (KAN-FCM) that replaces a portion of the traditional fully connected classifier. This module employs piecewise polynomial (spline) approximation to achieve adaptive nonlinear mapping while maintaining a controlled parameter budget. We evaluate EKAResNet on CIFAR-10 and CIFAR-100, achieving top accuracies of 95.84% and 80.06%, respectively. Importantly, the model maintains a parameter count comparable to strong ResNet and WideResNet baselines. Ablation studies on spline configurations further confirm the contribution of the KAN head. These results demonstrate the effectiveness of integrating KAN structures into ResNet for modeling high-dimensional, complex features. Our work highlights a promising direction for designing deep learning architectures that balance accuracy and computational efficiency.<\/jats:p>","DOI":"10.3390\/computation13110248","type":"journal-article","created":{"date-parts":[[2025,10,23]],"date-time":"2025-10-23T02:03:48Z","timestamp":1761185028000},"page":"248","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["EKAResNet: Enhancing ResNet with Kolmogorov\u2013Arnold Network-Based Nonlinear Feature Mapping"],"prefix":"10.3390","volume":"13","author":[{"ORCID":"https:\/\/orcid.org\/0009-0005-8242-8493","authenticated-orcid":false,"given":"Zhiming","family":"Dang","sequence":"first","affiliation":[{"name":"Department of Physics and Electrical Engineering, Mudanjiang Normal University, Mudanjiang 157000, China"}]},{"given":"Tonghua","family":"Wu","sequence":"additional","affiliation":[{"name":"Department of Physics and Electrical Engineering, Mudanjiang Normal University, Mudanjiang 157000, China"}]},{"given":"Wulin","family":"Zhang","sequence":"additional","affiliation":[{"name":"Department of Physics and Electrical Engineering, Mudanjiang Normal University, Mudanjiang 157000, China"}]},{"given":"Jianxin","family":"Chen","sequence":"additional","affiliation":[{"name":"Department of Physics and Electrical Engineering, Mudanjiang Normal University, Mudanjiang 157000, China"}]},{"given":"Huanlin","family":"Chen","sequence":"additional","affiliation":[{"name":"Department of Physics and Electrical Engineering, Mudanjiang Normal University, Mudanjiang 157000, China"}]},{"given":"Xuan","family":"Liu","sequence":"additional","affiliation":[{"name":"Department of Physics and Electrical Engineering, Mudanjiang Normal University, Mudanjiang 157000, China"}]},{"given":"Zirui","family":"Liu","sequence":"additional","affiliation":[{"name":"Department of Physics and Electrical Engineering, Mudanjiang Normal University, Mudanjiang 157000, China"}]}],"member":"1968","published-online":{"date-parts":[[2025,10,22]]},"reference":[{"key":"ref_1","unstructured":"He, K., Zhang, X., Ren, S., and Sun, J. (July, January 26). Deep Residual Learning for Image Recognition. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA."},{"key":"ref_2","first-page":"249","article-title":"Understanding the Difficulty of Training Deep Feedforward Neural Networks","volume":"Volume 9","author":"Glorot","year":"2010","journal-title":"Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics, Chia Laguna Resort"},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"7068349","DOI":"10.1155\/2018\/7068349","article-title":"Deep learning for computer vision: A brief review","volume":"2018","author":"Voulodimos","year":"2018","journal-title":"Comput. Intell. Neurosci."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"107","DOI":"10.1142\/S0218488598000094","article-title":"The vanishing gradient problem during learning recurrent neural nets and problem solutions","volume":"6","author":"Hochreiter","year":"1998","journal-title":"Int. J. Uncertain. Fuzziness Knowl.-Based Syst."},{"key":"ref_5","doi-asserted-by":"crossref","unstructured":"Botalb, A., Moinuddin, M., Al-Saggaf, U.M., and Ali, S.S. (2018, January 23\u201325). Contrasting convolutional neural network (CNN) with multi-layer perceptron (MLP) for big data analysis. Proceedings of the 2018 International Conference on Intelligent and Advanced System (ICIAS), Kuala Lumpur, Malaysia.","DOI":"10.1109\/ICIAS.2018.8540626"},{"key":"ref_6","doi-asserted-by":"crossref","unstructured":"Krichen, M. (2023). Convolutional neural networks: A survey. Computers, 12.","DOI":"10.3390\/computers12080151"},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"Borawar, L., and Kaur, R. (2022, January 9\u201310). ResNet: Solving vanishing gradient in deep networks. Proceedings of the International Conference on Recent Trends in Computing (ICRTC 2022), Chandigarh, India.","DOI":"10.1007\/978-981-19-8825-7_21"},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Cao, Y., Li, X., Feng, H., Wang, T., Zhang, J., and Liu, J. (2024, January 1\u20133). KANMixer: A Fully Linear Network with Multi-Scale Feature Awareness for Power Quality Classification Tasks. Proceedings of the 2024 China Automation Congress (CAC), Xiamen, China.","DOI":"10.1109\/CAC63892.2024.10865384"},{"key":"ref_9","doi-asserted-by":"crossref","unstructured":"Reinhardt, E., Ramakrishnan, D., and Gleyzer, S. (2025). Sinekan: Kolmogorov-arnold networks using sinusoidal activation functions. Front. Artif. Intell., 7.","DOI":"10.3389\/frai.2024.1462952"},{"key":"ref_10","unstructured":"Bodner, A.D., Tepsich, A.S., Spolski, J.N., and Pourteau, S. (2024). Convolutional kolmogorov-arnold networks. arXiv."},{"key":"ref_11","unstructured":"Zheng, L.N., Zhang, W.E., Yue, L., Xu, M., Maennel, O., and Chen, W. (2025). Free-Knots Kolmogorov-Arnold Network: On the Analysis of Spline Knots and Advancing Stability. arXiv."},{"key":"ref_12","unstructured":"Ji, T., Hou, Y., and Zhang, D. (2024). A comprehensive survey on Kolmogorov Arnold networks (KAN). arXiv."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"128585","DOI":"10.1016\/j.physa.2023.128585","article-title":"Structure and performance of fully connected neural networks: Emerging complex network properties","volume":"615","author":"Scabini","year":"2023","journal-title":"Phys. A Stat. Mech. Its Appl."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"16817","DOI":"10.1007\/s11042-022-13995-6","article-title":"Improving deep learning-based polyp detection using feature extraction and data augmentation","volume":"82","author":"Chou","year":"2023","journal-title":"Multimed. Tools Appl."},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"Chen, W., Wang, Y., Ren, Y., Jiang, H., Du, G., Zhang, J., and Li, J. (2023). An automated detection of epileptic seizures EEG using CNN classifier based on feature fusion with high accuracy. BMC Med. Inform. Decis. Mak., 23.","DOI":"10.1186\/s12911-023-02180-w"},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"367","DOI":"10.1080\/05704928.2021.1999252","article-title":"An overview of hyperspectral image feature extraction, classification methods and the methods based on small samples","volume":"58","author":"Li","year":"2023","journal-title":"Appl. Spectrosc. Rev."},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"79","DOI":"10.1007\/s10916-022-01863-7","article-title":"Human monkeypox classification from skin lesion images with deep pre-trained network using mobile application","volume":"46","author":"Sahin","year":"2022","journal-title":"J. Med. Syst."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"120842","DOI":"10.1016\/j.eswa.2023.120842","article-title":"ReX-Net: A reflectance-guided underwater image enhancement network for extreme scenarios","volume":"231","author":"Zhang","year":"2023","journal-title":"Expert Syst. Appl."},{"key":"ref_19","doi-asserted-by":"crossref","unstructured":"Zhao, L., and Zhang, Z. (2024). A improved pooling method for convolutional neural networks. Sci. Rep., 14.","DOI":"10.1038\/s41598-024-51258-6"},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"4516","DOI":"10.1109\/TFUZZ.2023.3287193","article-title":"A robust multilabel feature selection approach based on graph structure considering fuzzy dependency and feature interaction","volume":"31","author":"Yin","year":"2023","journal-title":"IEEE Trans. Fuzzy Syst."},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Zagoruyko, S., and Komodakis, N. (2016). Wide residual networks. arXiv.","DOI":"10.5244\/C.30.87"},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"101567","DOI":"10.1016\/j.jksuci.2023.101567","article-title":"DARGS: Image inpainting algorithm via deep attention residuals group and semantics","volume":"35","author":"Chen","year":"2023","journal-title":"J. King Saud Univ. Comput. Inf. Sci."},{"key":"ref_23","doi-asserted-by":"crossref","unstructured":"Cao, L., Li, J., and Chen, S. (2023). Multi-target segmentation of pancreas and pancreatic tumor based on fusion of attention mechanism. Biomed. Signal Process. Control, 79.","DOI":"10.1016\/j.bspc.2022.104170"},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"3625","DOI":"10.1109\/TCSI.2023.3287602","article-title":"An efficient CNN inference accelerator based on intra-and inter-channel feature map compression","volume":"70","author":"Xie","year":"2023","journal-title":"IEEE Trans. Circuits Syst. Regul. Pap."},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Mou, L., Xiao, X., Cao, W., Li, W., and Chen, X. (2024, January 18\u201320). Efficient and accurate capsule networks with b-spline-based activation functions. Proceedings of the 2024 International Conference on New Trends in Computational Intelligence (NTCI), Shanghai, China.","DOI":"10.1109\/NTCI64025.2024.10776434"},{"key":"ref_26","doi-asserted-by":"crossref","unstructured":"Ta, H.T. (2024). BSRBF-KAN: A combination of B-splines and Radial Basis Functions in Kolmogorov-Arnold Networks. arXiv.","DOI":"10.1007\/978-981-96-4288-5_1"},{"key":"ref_27","unstructured":"Liu, Z., Wang, Y., Vaidya, S., Ruehle, F., Halverson, J., Solja\u010di\u0107, M., and Tegmark, M. (2024). KAN: Kolmogorov-arnold networks. arXiv."},{"key":"ref_28","unstructured":"Cheon, M. (2024). Kolmogorov-arnold network for satellite image classification in remote sensing. arXiv."},{"key":"ref_29","first-page":"1","article-title":"Rsmamba: Remote sensing image classification with state space model","volume":"21","author":"Chen","year":"2024","journal-title":"IEEE Geosci. Remote Sens. Lett."},{"key":"ref_30","unstructured":"Yu, R.C., Wu, S., and Gui, J. (2024). Residual Kolmogorov\u2013Arnold Network for Enhanced Deep Learning. arXiv."},{"key":"ref_31","unstructured":"Yang, X., and Wang, X. (2024). Kolmogorov\u2013Arnold Transformer. arXiv."},{"key":"ref_32","doi-asserted-by":"crossref","unstructured":"Woo, S., Kim, D., Park, J., Lee, J., and Kweon, I.S. (2023, January 18\u201322). ConvNeXt V2: Co-Designing and Scaling ConvNets with Masked Autoencoders. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Vancouver, BC, Canada.","DOI":"10.1109\/CVPR52729.2023.01548"}],"container-title":["Computation"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2079-3197\/13\/11\/248\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,23]],"date-time":"2025-10-23T02:30:11Z","timestamp":1761186611000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2079-3197\/13\/11\/248"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,22]]},"references-count":32,"journal-issue":{"issue":"11","published-online":{"date-parts":[[2025,11]]}},"alternative-id":["computation13110248"],"URL":"https:\/\/doi.org\/10.3390\/computation13110248","relation":{},"ISSN":["2079-3197"],"issn-type":[{"value":"2079-3197","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,10,22]]}}}