{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,11,24]],"date-time":"2025-11-24T07:14:32Z","timestamp":1763968472713,"version":"3.40.1"},"reference-count":82,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"11","license":[{"start":{"date-parts":[[2022,11,1]],"date-time":"2022-11-01T00:00:00Z","timestamp":1667260800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2022,11,1]],"date-time":"2022-11-01T00:00:00Z","timestamp":1667260800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2022,11,1]],"date-time":"2022-11-01T00:00:00Z","timestamp":1667260800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["61632003","62072449"],"award-info":[{"award-number":["61632003","62072449"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Macau FDCT","award":["0018\/2019\/AKP","SKL-IOTSC-2021-2023"],"award-info":[{"award-number":["0018\/2019\/AKP","SKL-IOTSC-2021-2023"]}]},{"DOI":"10.13039\/501100000923","name":"Australian Research Council","doi-asserted-by":"publisher","award":["DE180101438","DP210101859"],"award-info":[{"award-number":["DE180101438","DP210101859"]}],"id":[{"id":"10.13039\/501100000923","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Pattern Anal. Mach. Intell."],"published-print":{"date-parts":[[2022,11,1]]},"DOI":"10.1109\/tpami.2021.3114368","type":"journal-article","created":{"date-parts":[[2021,9,21]],"date-time":"2021-09-21T20:22:05Z","timestamp":1632255725000},"page":"7731-7746","source":"Crossref","is-referenced-by-count":7,"title":["Learning Versatile Convolution Filters for Efficient Visual Recognition"],"prefix":"10.1109","volume":"44","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-9761-2702","authenticated-orcid":false,"given":"Kai","family":"Han","sequence":"first","affiliation":[{"name":"State Key Laboratory of Computer Science, Institute of Software, Chinese Academy of Sciences, Beijing, China"}]},{"given":"Yunhe","family":"Wang","sequence":"additional","affiliation":[{"name":"Noah&#x0027;s Ark Lab, Huawei Technologies, Shenzhen, Guangdong, China"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4756-0609","authenticated-orcid":false,"given":"Chang","family":"Xu","sequence":"additional","affiliation":[{"name":"School of Computer Science, Faculty of Engineering, University of Sydney, Darlington, NSW, Australia"}]},{"given":"Chunjing","family":"Xu","sequence":"additional","affiliation":[{"name":"Noah&#x0027;s Ark Lab, Huawei Technologies, Shenzhen, Guangdong, China"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2174-1428","authenticated-orcid":false,"given":"Enhua","family":"Wu","sequence":"additional","affiliation":[{"name":"State Key Laboratory of Computer Science, Institute of Software, Chinese Academy of Sciences, Beijing, China"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5979-578X","authenticated-orcid":false,"given":"Dacheng","family":"Tao","sequence":"additional","affiliation":[{"name":"School of Computer Science, Faculty of Engineering, University of Sydney, Darlington, NSW, Australia"}]}],"member":"263","reference":[{"key":"ref73","first-page":"5987","article-title":"Aggregated residual transformations for deep neural networks","author":"xie","year":"2017","journal-title":"Proc IEEE Conf Comput Vis Pattern Recognit"},{"key":"ref72","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.521"},{"article-title":"Mixed precision quantization of convnets via differentiable neural architecture search","year":"2018","author":"wu","key":"ref71"},{"key":"ref70","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00951"},{"article-title":"Searching for low-bit weights in quantized neural networks","year":"2020","author":"yang","key":"ref76"},{"key":"ref77","article-title":"Slimmable neural networks","author":"yu","year":"2019","journal-title":"Proc Int Conf Learn Representations"},{"key":"ref74","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00748"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00290"},{"key":"ref75","first-page":"1","article-title":"FSNet: Compression of deep convolutional neural networks by filter summary","author":"yang","year":"2020","journal-title":"Proc Int Conf Learn Representations"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.00160"},{"key":"ref78","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00958"},{"key":"ref79","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01237-3_23"},{"article-title":"Learning multiple layers of features from tiny images","year":"2009","author":"krizhevsky","key":"ref33"},{"key":"ref32","first-page":"1","article-title":"Compression of deep convolutional neural networks for fast and low power mobile applications","author":"kim","year":"2016","journal-title":"Proc Int Conf Learn Representations"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.182"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1145\/2647868.2654889"},{"key":"ref37","first-page":"1","article-title":"Pruning filters for efficient convnets","author":"li","year":"2017","journal-title":"Proc Int Conf Learn Representations"},{"key":"ref36","first-page":"1","article-title":"Speeding-up convolutional neural networks using fine-tuned CP-decomposition","author":"lebedev","year":"2015","journal-title":"Proc Int Conf Learn Representations"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.435"},{"key":"ref34","first-page":"1097","article-title":"Imagenet classification with deep convolutional neural networks","author":"krizhevsky","year":"2012","journal-title":"Proc Int Conf Neural Inf Process"},{"key":"ref60","first-page":"6105","article-title":"Efficientnet: Rethinking model scaling for convolutional neural networks","author":"tan","year":"2019","journal-title":"Proc Int Conf Mach Learn"},{"key":"ref62","first-page":"10936","article-title":"SCOP: Scientific control for reliable neural network pruning","author":"tang","year":"2020","journal-title":"Proc Int Conf Neural Inf Process"},{"key":"ref61","first-page":"1","article-title":"Mixnet: Mixed depthwise convolutional kernels","author":"tan","year":"2019","journal-title":"Proc Brit Mach Vis Conf"},{"key":"ref63","first-page":"1","article-title":"Improving the speed of neural networks on CPUs","author":"vanhoucke","year":"2011","journal-title":"Proc Deep Learn Unsupervised Feature Learn Workshop Int Conf Neural Inf Process Syst"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.5244\/C.28.88"},{"key":"ref64","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.01298"},{"article-title":"SqueezeNet: Alexnet-level accuracy with 50x fewer parameters and< 0.5 MB model size","year":"2016","author":"iandola","key":"ref27"},{"key":"ref65","first-page":"3703","article-title":"Beyond filters: Compact feature map for portable deep model","author":"wang","year":"2017","journal-title":"Proc Int Conf Mach Learn"},{"key":"ref66","first-page":"1613","article-title":"Learning versatile filters for efficient convolutional neural networks","author":"wang","year":"2018","journal-title":"Proc Int Conf Neural Inf Process"},{"key":"ref29","first-page":"5955","article-title":"Constructing fast network through deconstruction of convolution","author":"jeon","year":"2018","journal-title":"Proc Int Conf Neural Inf Process"},{"key":"ref67","first-page":"253","article-title":"CNNpack: Packing convolutional neural networks in the frequency domain","author":"wang","year":"2016","journal-title":"Proc Int Conf Neural Inf Process"},{"key":"ref68","first-page":"2074","article-title":"Learning structured sparsity in deep neural networks","author":"wen","year":"2016","journal-title":"Proc Int Conf Neural Inf Process"},{"key":"ref69","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.01099"},{"key":"ref2","first-page":"584","article-title":"Provable bounds for learning some deep representations","author":"arora","year":"2014","journal-title":"Proc 31st Int Conf Int Conf Mach Learn"},{"key":"ref1","first-page":"265","article-title":"TensorFlow: A system for large-scale machine learning","author":"abadi","year":"2016"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"article-title":"Distilling the knowledge in a neural network","year":"2015","author":"hinton","key":"ref22"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.155"},{"article-title":"Mobilenets: Efficient convolutional neural networks for mobile vision applications","year":"2017","author":"howard","key":"ref24"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00140"},{"key":"ref26","first-page":"4107","article-title":"Binarized neural networks","author":"hubara","year":"2016","journal-title":"Proc Int Conf Neural Inf Process"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01270-0_19"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-46493-0_32"},{"key":"ref51","first-page":"91","article-title":"Faster R-CNN: Towards real-time object detection with region proposal networks","author":"ren","year":"2015","journal-title":"Proc Int Conf Neural Inf Process"},{"key":"ref59","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00293"},{"key":"ref58","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.308"},{"key":"ref57","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2015.7298594"},{"article-title":"Very deep convolutional networks for large-scale image recognition","year":"2014","author":"simonyan","key":"ref56"},{"key":"ref55","first-page":"1","article-title":"Learning implicitly recurrent CNNs through parameter sharing","author":"savarese","year":"2019","journal-title":"Proc Int Conf Learn Representations"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00474"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-015-0816-y"},{"key":"ref52","first-page":"1","article-title":"Fitnets: Hints for thin deep nets","author":"romero","year":"2015","journal-title":"Proc Int Conf Learn Representations"},{"article-title":"Binarized neural networks: Training deep neural networks with weights and activations constrained to+ 1 or-1","year":"2016","author":"courbariaux","key":"ref10"},{"key":"ref11","first-page":"1269","article-title":"Exploiting linear structure within convolutional networks for efficient evaluation","author":"denton","year":"2014","journal-title":"Proc Int Conf Neural Inf Process"},{"key":"ref40","first-page":"740","article-title":"Microsoft COCO: Common objects in context","author":"lin","year":"2014","journal-title":"Proc Eur Conf Comput Vis"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2015.2439281"},{"key":"ref13","first-page":"947","article-title":"PerforatedCNNs: Acceleration through elimination of redundant convolutions","author":"figurnov","year":"2016","journal-title":"Proc 30th Conf Neural Inf Process Syst"},{"key":"ref14","first-page":"5203","article-title":"ChannelNets: Compact and efficient convolutional neural networks via channel-wise convolutions","author":"gao","year":"2018","journal-title":"Proc Int Conf Neural Inf Process"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58517-4_32"},{"key":"ref82","first-page":"1","article-title":"Neural architecture search with reinforcement learning","author":"zoph","year":"2017","journal-title":"Proc Int Conf Learn Representations"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.00165"},{"key":"ref81","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00716"},{"key":"ref17","first-page":"4017","article-title":"Training binary neural networks through learning with noisy supervision","author":"han","year":"2020","journal-title":"Proc Int Conf Mach Learn"},{"key":"ref18","first-page":"1","article-title":"Deep compression: Compressing deep neural networks with pruning, trained quantization and Huffman coding","author":"han","year":"2016","journal-title":"Proc Int Conf Learn Representations"},{"key":"ref19","first-page":"1135","article-title":"Learning both weights and connections for efficient neural network","author":"han","year":"2015","journal-title":"Proc Int Conf Neural Inf Process"},{"key":"ref80","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.469"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.5244\/C.26.135"},{"article-title":"Estimating or propagating gradients through stochastic neurons for conditional computation","year":"2013","author":"bengio","key":"ref3"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00361"},{"key":"ref5","first-page":"1","article-title":"ProxylessNAS: Direct neural architecture search on target task and hardware","author":"cai","year":"2019","journal-title":"Proc Int Conf Learn Representations"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.195"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00353"},{"year":"2017","key":"ref49","article-title":"Ncnn"},{"key":"ref9","first-page":"3123","article-title":"BinaryConnect: Training deep neural networks with binary weights during propagations","author":"courbariaux","year":"2015","journal-title":"Proc Int Conf Neural Inf Process"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.541"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2015.7298965"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.01152"},{"key":"ref47","first-page":"122","article-title":"Shufflenet v2: Practical guidelines for efficient CNN architecture design","author":"ma","year":"2018","journal-title":"Proc Eur Conf Comput Vis"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.298"},{"key":"ref41","first-page":"806","article-title":"Sparse convolutional neural networks","author":"liu","year":"2015","journal-title":"Proc IEEE Conf Comput Vis Pattern Recognit"},{"key":"ref44","first-page":"747","article-title":"Bi-real net: Enhancing the performance of 1-bit CNNs with improved representational capability and advanced training algorithm","author":"liu","year":"2018","journal-title":"Proc Eur Conf Comput Vis"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00339"}],"container-title":["IEEE Transactions on Pattern Analysis and Machine Intelligence"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/34\/9910240\/09543586.pdf?arnumber=9543586","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,3,19]],"date-time":"2025-03-19T18:41:14Z","timestamp":1742409674000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/9543586\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,11,1]]},"references-count":82,"journal-issue":{"issue":"11"},"URL":"https:\/\/doi.org\/10.1109\/tpami.2021.3114368","relation":{},"ISSN":["0162-8828","2160-9292","1939-3539"],"issn-type":[{"type":"print","value":"0162-8828"},{"type":"electronic","value":"2160-9292"},{"type":"electronic","value":"1939-3539"}],"subject":[],"published":{"date-parts":[[2022,11,1]]}}}