{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,30]],"date-time":"2026-04-30T14:40:30Z","timestamp":1777560030574,"version":"3.51.4"},"reference-count":22,"publisher":"SAGE Publications","issue":"4","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["AIC"],"published-print":{"date-parts":[[2023,10,13]]},"abstract":"<jats:p>Pruning of neural networks is undoubtedly a popular approach to cope with the current compression of large-scale, high-cost network models. However, most of the existing methods require a high level of human-regulated pruning criteria, which requires a lot of human effort to figure out a reasonable pruning strength. One of the main reasons is that there are different levels of sensitivity distribution in the network. Our main goal is to discover compression methods that adapt to this distribution to avoid deep architectural damage to the network due to unnecessary pruning. In this paper, we propose a filter texture distribution that affects the training of the network. We also analyze the sensitivity of each of the diverse states of this distribution. To do so, we first use a multidimensional penalty method that can analyze the potential sensitivity based on this texture distribution to obtain a pruning-friendly sparse environment. Then, we set up a lightweight dynamic threshold container in order to prune the sparse network. By providing each filter with a suitable threshold for that filter at a low cost, a massive reduction in the number of parameters is achieved without affecting the contribution of certain pruning-sensitive layers to the network as a whole. In the final experiments, our two methods adapted to texture distribution were applied to ResNet Deep Neural Network (DNN) and VGG-16, which were deployed on the classical CIFAR-10\/100 and ImageNet datasets with excellent results in order to facilitate comparison with good cutting-edge pruning methods. Code is available at https:\/\/github.com\/wangyuzhe27\/CDP-and-DTC.<\/jats:p>","DOI":"10.3233\/aic-230046","type":"journal-article","created":{"date-parts":[[2023,8,25]],"date-time":"2023-08-25T10:29:47Z","timestamp":1692959387000},"page":"311-323","source":"Crossref","is-referenced-by-count":0,"title":["Dynamic finegrained structured pruning sensitive to filter texture distribution"],"prefix":"10.1177","volume":"36","author":[{"given":"Ping","family":"Li","sequence":"first","affiliation":[{"name":"School of Computer and Communication Engineering, Changsha University of Science and Technology, Hunan, China"},{"name":"Hunan Provincial Key Laboratory of Intelligent Processing of Big Data on Transportation, Hunan, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yuzhe","family":"Wang","sequence":"additional","affiliation":[{"name":"School of Computer and Communication Engineering, Changsha University of Science and Technology, Hunan, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Cong","family":"Wu","sequence":"additional","affiliation":[{"name":"School of Computer and Communication Engineering, Changsha University of Science and Technology, Hunan, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xiatao","family":"Kang","sequence":"additional","affiliation":[{"name":"School of Computer and Communication Engineering, Changsha University of Science and Technology, Hunan, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"179","reference":[{"key":"10.3233\/AIC-230046_ref4","doi-asserted-by":"crossref","unstructured":"X.\u00a0Ding, Y.\u00a0Guo, G.\u00a0Ding and J.\u00a0Han, ACNet: Strengthening the kernel skeletons for powerful CNN via asymmetric convolution blocks, in: Proceedings of the IEEE\/CVF International Conference on Computer Vision, 2019, pp.\u00a01911\u20131920.","DOI":"10.1109\/ICCV.2019.00200"},{"key":"10.3233\/AIC-230046_ref5","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2013.6638947"},{"issue":"3","key":"10.3233\/AIC-230046_ref6","doi-asserted-by":"publisher","first-page":"243","DOI":"10.1145\/3007787.3001163","article-title":"EIE: Efficient inference engine on compressed deep neural network","volume":"44","author":"Han","year":"2016","journal-title":"ACM SIGARCH Computer Architecture News"},{"key":"10.3233\/AIC-230046_ref9","doi-asserted-by":"crossref","unstructured":"Y.\u00a0He, Y.\u00a0Ding, P.\u00a0Liu, L.\u00a0Zhu, H.\u00a0Zhang and Y.\u00a0Yang, Learning filter pruning criteria for deep convolutional neural networks acceleration, in: Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, 2020, pp.\u00a02009\u20132018.","DOI":"10.1109\/CVPR42600.2020.00208"},{"key":"10.3233\/AIC-230046_ref11","doi-asserted-by":"crossref","unstructured":"Y.\u00a0He, J.\u00a0Lin, Z.\u00a0Liu, H.\u00a0Wang, L.-J.\u00a0Li and S.\u00a0Han, Amc: Automl for model compression and acceleration on mobile devices, in: Proceedings of the European Conference on Computer Vision (ECCV), 2018, pp.\u00a0784\u2013800.","DOI":"10.1007\/978-3-030-01234-2_48"},{"key":"10.3233\/AIC-230046_ref12","doi-asserted-by":"crossref","unstructured":"Y.\u00a0He, P.\u00a0Liu, Z.\u00a0Wang, Z.\u00a0Hu and Y.\u00a0Yang, Filter pruning via geometric median for deep convolutional neural networks acceleration, in: Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, 2019, pp.\u00a04340\u20134349.","DOI":"10.1109\/CVPR.2019.00447"},{"key":"10.3233\/AIC-230046_ref14","doi-asserted-by":"crossref","unstructured":"Z.\u00a0Huang and N.\u00a0Wang, Data-driven sparse structure selection for deep neural networks, in: Proceedings of the European Conference on Computer Vision (ECCV), 2018, pp.\u00a0304\u2013320.","DOI":"10.1007\/978-3-030-01270-0_19"},{"issue":"6","key":"10.3233\/AIC-230046_ref15","doi-asserted-by":"publisher","first-page":"84","DOI":"10.1145\/3065386","article-title":"ImageNet classification with deep convolutional neural networks","volume":"60","author":"Krizhevsky","year":"2017","journal-title":"Communications of the ACM"},{"key":"10.3233\/AIC-230046_ref16","unstructured":"A.\u00a0Kusupati, V.\u00a0Ramanujan, R.\u00a0Somani, M.\u00a0Wortsman, P.\u00a0Jain, S.\u00a0Kakade and A.\u00a0Farhadi, Soft threshold weight reparameterization for learnable sparsity, in: International Conference on Machine Learning, PMLR, 2020, pp.\u00a05544\u20135555."},{"key":"10.3233\/AIC-230046_ref17","doi-asserted-by":"crossref","unstructured":"V.\u00a0Lebedev and V.\u00a0Lempitsky, Fast ConvNets using group-wise brain damage, in: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2016, pp.\u00a02554\u20132564.","DOI":"10.1109\/CVPR.2016.280"},{"key":"10.3233\/AIC-230046_ref21","doi-asserted-by":"crossref","unstructured":"M.\u00a0Lin, R.\u00a0Ji, Y.\u00a0Wang, Y.\u00a0Zhang, B.\u00a0Zhang, Y.\u00a0Tian and L.\u00a0Shao, Hrank: Filter pruning using high-rank feature map, in: Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, 2020, pp.\u00a01529\u20131538.","DOI":"10.1109\/CVPR42600.2020.00160"},{"key":"10.3233\/AIC-230046_ref22","doi-asserted-by":"crossref","unstructured":"S.\u00a0Lin, R.\u00a0Ji, C.\u00a0Yan, B.\u00a0Zhang, L.\u00a0Cao, Q.\u00a0Ye, F.\u00a0Huang and D.\u00a0Doermann, Towards optimal structured cnn pruning via generative adversarial learning, in: Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, 2019, pp.\u00a02790\u20132799.","DOI":"10.1109\/CVPR.2019.00290"},{"key":"10.3233\/AIC-230046_ref23","doi-asserted-by":"crossref","unstructured":"N.\u00a0Liu, X.\u00a0Ma, Z.\u00a0Xu, Y.\u00a0Wang, J.\u00a0Tang and J.\u00a0Ye, Autocompress: An automatic DNN structured pruning framework for ultra-high compression rates, in: Proceedings of the AAAI Conference on Artificial Intelligence, Vol.\u00a034, 2020, pp.\u00a04876\u20134883.","DOI":"10.1609\/aaai.v34i04.5924"},{"key":"10.3233\/AIC-230046_ref24","doi-asserted-by":"crossref","unstructured":"Z.\u00a0Liu, J.\u00a0Li, Z.\u00a0Shen, G.\u00a0Huang, S.\u00a0Yan and C.\u00a0Zhang, Learning efficient convolutional networks through network slimming, in: Proceedings of the IEEE International Conference on Computer Vision, 2017, pp.\u00a02736\u20132744.","DOI":"10.1109\/ICCV.2017.298"},{"key":"10.3233\/AIC-230046_ref27","doi-asserted-by":"crossref","unstructured":"J.-H.\u00a0Luo, J.\u00a0Wu and W.\u00a0Lin, ThiNet: A filter level pruning method for deep neural network compression, in: Proceedings of the IEEE International Conference on Computer Vision, 2017, pp.\u00a05058\u20135066.","DOI":"10.1109\/ICCV.2017.541"},{"key":"10.3233\/AIC-230046_ref28","doi-asserted-by":"crossref","unstructured":"H.\u00a0Mao, S.\u00a0Han, J.\u00a0Pool, W.\u00a0Li, X.\u00a0Liu, Y.\u00a0Wang and W.J.\u00a0Dally, Exploring the granularity of sparsity in convolutional neural networks, in: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops, 2017, pp.\u00a013\u201320.","DOI":"10.1109\/CVPRW.2017.241"},{"key":"10.3233\/AIC-230046_ref29","first-page":"17629","article-title":"Pruning filter in filter","volume":"33","author":"Meng","year":"2020","journal-title":"Advances in Neural Information Processing Systems"},{"key":"10.3233\/AIC-230046_ref31","doi-asserted-by":"crossref","unstructured":"C.\u00a0Szegedy, V.\u00a0Vanhoucke, S.\u00a0Ioffe, J.\u00a0Shlens and Z.\u00a0Wojna, Rethinking the inception architecture for computer vision, in: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2016, pp.\u00a02818\u20132826.","DOI":"10.1109\/CVPR.2016.308"},{"key":"10.3233\/AIC-230046_ref34","unstructured":"H.\u00a0Wang, C.\u00a0Qin, Y.\u00a0Zhang and Y.\u00a0Fu, Neural pruning via growing regularization, in: International Conference on Learning Representations, 2021."},{"key":"10.3233\/AIC-230046_ref35","doi-asserted-by":"crossref","unstructured":"H.\u00a0Wang, Q.\u00a0Zhang, Y.\u00a0Wang, L.\u00a0Yu and H.\u00a0Hu, Structured pruning for efficient convnets via incremental regularization, in: 2019 International Joint Conference on Neural Networks (IJCNN), IEEE, 2019, pp.\u00a01\u20138.","DOI":"10.1109\/IJCNN.2019.8852463"},{"key":"10.3233\/AIC-230046_ref39","doi-asserted-by":"crossref","unstructured":"R.\u00a0Yu, A.\u00a0Li, C.-F.\u00a0Chen, J.-H.\u00a0Lai, V.I.\u00a0Morariu, X.\u00a0Han, M.\u00a0Gao, C.-Y.\u00a0Lin and L.S.\u00a0Davis, NISP: Pruning networks using neuron importance score propagation, in: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2018, pp.\u00a09194\u20139203.","DOI":"10.1109\/CVPR.2018.00958"},{"issue":"1","key":"10.3233\/AIC-230046_ref40","doi-asserted-by":"publisher","first-page":"49","DOI":"10.1111\/j.1467-9868.2005.00532.x","article-title":"Model selection and estimation in regression with grouped variables","volume":"68","author":"Yuan","year":"2006","journal-title":"Journal of the Royal Statistical Society: Series B (Statistical Methodology)"}],"container-title":["AI Communications"],"original-title":[],"link":[{"URL":"https:\/\/content.iospress.com\/download?id=10.3233\/AIC-230046","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,28]],"date-time":"2026-04-28T18:28:10Z","timestamp":1777400890000},"score":1,"resource":{"primary":{"URL":"https:\/\/journals.sagepub.com\/doi\/full\/10.3233\/AIC-230046"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,10,13]]},"references-count":22,"journal-issue":{"issue":"4"},"URL":"https:\/\/doi.org\/10.3233\/aic-230046","relation":{},"ISSN":["1875-8452","0921-7126"],"issn-type":[{"value":"1875-8452","type":"electronic"},{"value":"0921-7126","type":"print"}],"subject":[],"published":{"date-parts":[[2023,10,13]]}}}