{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,12,5]],"date-time":"2025-12-05T18:53:12Z","timestamp":1764960792640,"version":"3.46.0"},"reference-count":42,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"3","license":[{"start":{"date-parts":[[2025,3,1]],"date-time":"2025-03-01T00:00:00Z","timestamp":1740787200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2025,3,1]],"date-time":"2025-03-01T00:00:00Z","timestamp":1740787200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,3,1]],"date-time":"2025-03-01T00:00:00Z","timestamp":1740787200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"name":"National Key Research and Development Program of China","award":["2018AAA0101503"],"award-info":[{"award-number":["2018AAA0101503"]}]},{"DOI":"10.13039\/501100010880","name":"Science and Technology Project of State Grid Corporation of China (SGCC)","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100010880","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Neural Netw. Learning Syst."],"published-print":{"date-parts":[[2025,3]]},"DOI":"10.1109\/tnnls.2021.3106917","type":"journal-article","created":{"date-parts":[[2021,9,6]],"date-time":"2021-09-06T16:19:03Z","timestamp":1630945143000},"page":"3930-3942","source":"Crossref","is-referenced-by-count":10,"title":["Adding Before Pruning: Sparse Filter Fusion for Deep Convolutional Neural Networks via Auxiliary Attention"],"prefix":"10.1109","volume":"36","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-7292-4056","authenticated-orcid":false,"given":"Guanzhong","family":"Tian","sequence":"first","affiliation":[{"name":"Institute of Cyber-Systems and Control, Zhejiang University, Hangzhou, China"}]},{"given":"Yiran","family":"Sun","sequence":"additional","affiliation":[{"name":"Institute of Cyber-Systems and Control, Zhejiang University, Hangzhou, China"}]},{"given":"Yuang","family":"Liu","sequence":"additional","affiliation":[{"name":"Institute of Cyber-Systems and Control, Zhejiang University, Hangzhou, China"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1251-2129","authenticated-orcid":false,"given":"Xianfang","family":"Zeng","sequence":"additional","affiliation":[{"name":"Institute of Cyber-Systems and Control, Zhejiang University, Hangzhou, China"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4035-0630","authenticated-orcid":false,"given":"Mengmeng","family":"Wang","sequence":"additional","affiliation":[{"name":"Institute of Cyber-Systems and Control, Zhejiang University, Hangzhou, China"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4822-8939","authenticated-orcid":false,"given":"Yong","family":"Liu","sequence":"additional","affiliation":[{"name":"Institute of Cyber-Systems and Control, Zhejiang University, Hangzhou, China"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8891-6766","authenticated-orcid":false,"given":"Jiangning","family":"Zhang","sequence":"additional","affiliation":[{"name":"Institute of Cyber-Systems and Control, Zhejiang University, Hangzhou, China"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6568-8801","authenticated-orcid":false,"given":"Jun","family":"Chen","sequence":"additional","affiliation":[{"name":"Institute of Cyber-Systems and Control, Zhejiang University, Hangzhou, China"}]}],"member":"263","reference":[{"key":"ref1","first-page":"605","article-title":"Globally optimal gradient descent for a convnet with Gaussian inputs","volume-title":"Proc. Int. Conf. Mach. Learn. (ICML)","author":"Brutzkus"},{"key":"ref2","first-page":"1339","article-title":"Gradient descent learns one-hidden-layer CNN: Don\u2019t be afraid of spurious local minima","volume-title":"Proc. Int. Conf. Mach. Learn. (ICML)","author":"Du"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-014-0733-5"},{"key":"ref4","first-page":"1379","article-title":"Dynamic network surgery for efficient DNNs","volume-title":"Proc. Adv. Neural Inf. Process. Syst. (NIPS)","author":"Guo"},{"key":"ref5","first-page":"1","article-title":"Deep compression: Compressing deep neural networks with pruning, trained quantization and Huffman coding","volume-title":"Proc. Int. Conf. Learn. Represent. (ICLR)","author":"Han"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.24963\/ijcai.2018\/309"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00447"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.155"},{"issue":"1","key":"ref10","first-page":"2146","article-title":"Neural networks for machine learning","volume":"264","author":"Hinton","year":"2012","journal-title":"Coursera"},{"key":"ref11","first-page":"1","article-title":"Distilling the knowledge in a neural network","volume-title":"Proc. Neural Inf. Process. Systems (NIPS)","author":"Hinton"},{"key":"ref12","article-title":"Network trimming: A data-driven neuron pruning approach towards efficient deep architectures","volume-title":"arXiv:1607.03250","author":"Hu","year":"2016"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00745"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01270-0_19"},{"key":"ref15","first-page":"4107","article-title":"Binarized neural networks","volume-title":"Proc. Adv. Neural Inf. Process. Syst. (NIPS)","author":"Hubara"},{"key":"ref16","article-title":"Batch normalization: Accelerating deep network training by reducing internal covariate shift","volume-title":"arXiv:1502.03167","author":"Ioffe","year":"2015"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2017.2689098"},{"article-title":"Learning multiple layers of features from tiny images","year":"2009","author":"Krizhevsky","key":"ref18"},{"key":"ref19","first-page":"1","article-title":"Pruning filters for efficient convnets","volume-title":"Proc. Int. Conf. Learn. Represent. (ICLR)","author":"Li"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00410"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2020.3048630"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.00160"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00290"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.l007\/978-3-319-46448-0_2"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.298"},{"key":"ref26","first-page":"1","article-title":"Rethinking the value of network pruning","volume-title":"Proc. Int. Conf. Learn. Represent. (ICLR)","author":"Liu"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.541"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2020.3034487"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.2307\/j.ctvcm4g18.8"},{"key":"ref30","first-page":"1","article-title":"Very deep convolutional networks for large-scale image recognition","volume-title":"Proc. Int. Conf. Learn. Represent. (ICLR)","author":"Simonyan"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1007\/s10462-020-09854-1"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.683"},{"key":"ref33","article-title":"Dynamical isometry: The missing ingredient for neural network pruning","volume-title":"arXiv:2105.05916","author":"Wang","year":"2021"},{"key":"ref34","first-page":"3","article-title":"Structured probabilistic pruning for convolutional neural network acceleration","volume-title":"Proc. Brit. Mach. Vis. Conf. (BMVC)","author":"Wang"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1007\/s11432-020-2969-8"},{"key":"ref36","first-page":"1","article-title":"PCAS: Pruning channels with attention statistics for deep network compression","volume-title":"Proc. Brit. Mach. Vis. Conf. (BMVC)","author":"Yamamoto"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1016\/j.ins.2019.12.057"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00958"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01237-3_12"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00289"},{"key":"ref41","first-page":"4140","article-title":"Recovery guarantees for one-hidden-layer neural networks","volume-title":"Proc. Int. Conf. Mach. Learn. (ICML)","author":"Zhong"},{"key":"ref42","first-page":"875","article-title":"Discrimination-aware channel pruning for deep neural networks","volume-title":"Proc. Adv. Neural Inf. Process. Syst. (NIPS)","author":"Zhuang"}],"container-title":["IEEE Transactions on Neural Networks and Learning Systems"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/5962385\/10908444\/09530256.pdf?arnumber=9530256","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,12,5]],"date-time":"2025-12-05T18:39:13Z","timestamp":1764959953000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/9530256\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,3]]},"references-count":42,"journal-issue":{"issue":"3"},"URL":"https:\/\/doi.org\/10.1109\/tnnls.2021.3106917","relation":{},"ISSN":["2162-237X","2162-2388"],"issn-type":[{"type":"print","value":"2162-237X"},{"type":"electronic","value":"2162-2388"}],"subject":[],"published":{"date-parts":[[2025,3]]}}}