{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,2,21]],"date-time":"2025-02-21T20:20:22Z","timestamp":1740169222522,"version":"3.37.3"},"reference-count":73,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","license":[{"start":{"date-parts":[[2021,1,1]],"date-time":"2021-01-01T00:00:00Z","timestamp":1609459200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/legalcode"}],"funder":[{"DOI":"10.13039\/501100000266","name":"Engineering and Physical Sciences Research Council (EPSRC) Scholarship from the Neuroinformatics and Computational Neuroscience Doctoral Training Centre","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100000266","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100000266","name":"EPSRC Centre for Doctoral Training in Pervasive Parallelism","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100000266","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Huawei DDMPLab Innovation Research Grant"},{"name":"European Union\u2019s Horizon 2020 Research and Innovation Programme","award":["732204 (Bonseyes)"],"award-info":[{"award-number":["732204 (Bonseyes)"]}]},{"name":"Swiss State Secretariat for Education, Research and Innovation","award":["16.0159"],"award-info":[{"award-number":["16.0159"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Access"],"published-print":{"date-parts":[[2021]]},"DOI":"10.1109\/access.2021.3086321","type":"journal-article","created":{"date-parts":[[2021,6,4]],"date-time":"2021-06-04T19:59:29Z","timestamp":1622836769000},"page":"83199-83213","source":"Crossref","is-referenced-by-count":2,"title":["Substituting Convolutions for Neural Network Compression"],"prefix":"10.1109","volume":"9","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-5685-4724","authenticated-orcid":false,"given":"Elliot J.","family":"Crowley","sequence":"first","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9864-3666","authenticated-orcid":false,"given":"Gavin","family":"Gray","sequence":"additional","affiliation":[]},{"given":"Jack","family":"Turner","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8100-506X","authenticated-orcid":false,"given":"Amos","family":"Storkey","sequence":"additional","affiliation":[]}],"member":"263","reference":[{"key":"ref73","article-title":"BlockSwap: Fisher-guided block substitution for network compression on a budget","author":"turner","year":"2020","journal-title":"Proc Int Conf Learn Represent"},{"key":"ref72","article-title":"Improved regularization of convolutional neural networks with cutout","author":"devries","year":"2017","journal-title":"arXiv 1708 04552"},{"key":"ref71","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v33i01.33014780"},{"key":"ref70","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01246-5_2"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.634"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.633"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1109\/ASAP49362.2020.00039"},{"key":"ref32","article-title":"CuDNN: Efficient primitives for deep learning","author":"chetlur","year":"2014","journal-title":"Arxiv 1410 0759"},{"key":"ref31","first-page":"1","article-title":"ImageNet classification with deep convolutional neural networks","author":"krizhevsky","year":"2012","journal-title":"Proc Adv Neural Inf Process Syst"},{"key":"ref30","article-title":"ACDC: A structured efficient linear layer","author":"moczulski","year":"2016","journal-title":"Proc Int Conf Learn Represent"},{"key":"ref37","article-title":"SqueezeNet: AlexNet-level accuracy with 50&#x00D7; fewer parameters and <0.5 MB model size","author":"iandola","year":"2016","journal-title":"arXiv 1602 07360"},{"key":"ref36","first-page":"448","article-title":"Batch normalization: Accelerating deep network training by reducing internal covariate shift","author":"ioffe","year":"2015","journal-title":"Proc Int Conf Mach Learn"},{"key":"ref35","article-title":"MobileNets: Efficient convolutional neural networks for mobile vision applications","author":"howard","year":"2017","journal-title":"arXiv 1704 04861"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.195"},{"key":"ref60","doi-asserted-by":"publisher","DOI":"10.1137\/090752286"},{"key":"ref62","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.01156"},{"key":"ref61","article-title":"Compression of deep convolutional neural networks for fast and low power mobile applications","author":"kim","year":"2015","journal-title":"arXiv 1511 06530"},{"journal-title":"Machine Learning A Probabilistic Perspective","year":"2012","author":"murphy","key":"ref63"},{"key":"ref28","first-page":"2285","article-title":"Compressing neural networks with the hashing trick","author":"chen","year":"2015","journal-title":"Proc Int Conf Mach Learn"},{"key":"ref64","article-title":"Multi-scale context aggregation by dilated convolutions","author":"yu","year":"2016","journal-title":"Proc Int Conf Learn Represent"},{"key":"ref27","first-page":"2888","article-title":"Moonshine: Distilling with cheap convolutions","author":"crowley","year":"2018","journal-title":"Proc Adv Neural Inf Process Syst"},{"article-title":"Learning multiple layers of features from tiny images","year":"2009","author":"krizhevsky","key":"ref65"},{"key":"ref66","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-015-0816-y"},{"key":"ref29","first-page":"442","article-title":"Tensorizing neural networks","author":"novikov","year":"2015","journal-title":"Proc Adv Neural Inf Process Syst"},{"key":"ref67","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-46493-0_39"},{"key":"ref68","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2015.7298594"},{"key":"ref69","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00907"},{"journal-title":"Why the Future of Machine Learning is Tiny","year":"2018","author":"warden","key":"ref2"},{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1038\/nature14539"},{"key":"ref20","article-title":"Towards the limit of network quantization","author":"choi","year":"2017","journal-title":"Proc Int Conf Learn Represent"},{"key":"ref22","article-title":"Neural architecture search with reinforcement learning","author":"zoph","year":"2017","journal-title":"Proc Int Conf Learn Represent"},{"key":"ref21","article-title":"Striving for simplicity: The all convolutional net","author":"springenberg","year":"2014","journal-title":"arXiv 1412 6806"},{"key":"ref24","first-page":"2654","article-title":"Do deep nets really need to be deep?","author":"ba","year":"2014","journal-title":"Proc Adv Neural Inf Process Syst"},{"key":"ref23","article-title":"DARTS: Differentiable architecture search","author":"liu","year":"2019","journal-title":"Proc Int Conf Learn Represent"},{"article-title":"Rigid-motion scattering for image classification","year":"2014","author":"sifre","key":"ref26"},{"key":"ref25","article-title":"Distilling the knowledge in a neural network","author":"hinton","year":"2015","journal-title":"ArXiv 1503 02531"},{"key":"ref50","article-title":"Unifying distillation and privileged information","author":"lopez-paz","year":"2015","journal-title":"arXiv 1511 03643"},{"key":"ref51","first-page":"2023","article-title":"Learning using privileged information: Similarity control and knowledge transfer","volume":"16","author":"vapnik","year":"2015","journal-title":"J Mach Learn Res"},{"key":"ref59","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00474"},{"key":"ref58","doi-asserted-by":"publisher","DOI":"10.5244\/C.30.87"},{"key":"ref57","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-46493-0_38"},{"key":"ref56","article-title":"Paying more attention to attention: Improving the performance of convolutional neural networks via attention transfer","author":"zagoruyko","year":"2017","journal-title":"Proc Int Conf Learn Represent"},{"key":"ref55","article-title":"Do deep convolutional nets really need to be deep and convolutional?","author":"urban","year":"2017","journal-title":"Proc Int Conf Learn Represent"},{"key":"ref54","article-title":"FitNets: Hints for thin deep nets","author":"romero","year":"2015","journal-title":"Proc Int Conf Learn Represent"},{"key":"ref53","article-title":"Net2Net: Accelerating learning via knowledge transfer","author":"chen","year":"2015","journal-title":"arXiv 1511 05641"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00454"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00137"},{"key":"ref11","first-page":"598","article-title":"Optimal brain damage","author":"lecun","year":"1989","journal-title":"Proc Adv Neural Inf Process Syst"},{"key":"ref40","article-title":"Design of efficient convolutional layers using single intra-channel convolution, topological subdivisioning and spatial &#x2018;Bottleneck&#x2019; structure","author":"wang","year":"2016","journal-title":"arXiv 1608 04337"},{"key":"ref12","first-page":"1135","article-title":"Learning both weights and connections for efficient neural networks","author":"han","year":"2015","journal-title":"Proc Adv Neural Inf Process Syst"},{"key":"ref13","article-title":"Pruning convolutional neural networks for resource efficient inference","author":"molchanov","year":"2017","journal-title":"Proc Int Conf Learn Represent"},{"key":"ref14","article-title":"Faster gaze prediction with dense networks and Fisher pruning","author":"theis","year":"2018","journal-title":"arXiv 1801 05787"},{"key":"ref15","article-title":"The lottery ticket hypothesis: Finding sparse, trainable neural networks","author":"frankle","year":"2019","journal-title":"Proc Int Conf Learn Represent"},{"key":"ref16","article-title":"Binarized neural networks: Training deep neural networks with weights and activations constrained to +1 or ?1","author":"courbariaux","year":"2016","journal-title":"arXiv 1602 02830 [cs]"},{"key":"ref17","article-title":"Trained ternary quantization","author":"zhu","year":"2017","journal-title":"Proc Int Conf Learn Represent"},{"key":"ref18","first-page":"6869","article-title":"Quantized neural networks: Training neural networks with low precision weights and activations","volume":"18","author":"hubara","year":"2017","journal-title":"J Mach Learn Res"},{"key":"ref19","article-title":"Incremental network quantization: Towards lossless CNNs with low-precision weights","author":"zhou","year":"2017","journal-title":"Proc Int Conf Learn Represent"},{"key":"ref4","article-title":"Very deep convolutional networks for large-scale image recognition","author":"simonyan","year":"2015","journal-title":"Proc Int Conf Learn Represent"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1145\/3161174"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.308"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref8","first-page":"4095","article-title":"Efficient neural architecture search via parameters sharing","author":"pham","year":"2018","journal-title":"Proc Int Conf Mach Learn"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.243"},{"key":"ref49","first-page":"3438","article-title":"Bayesian dark knowledge","author":"korattikara","year":"2015","journal-title":"Proc Adv Neural Inf Process Syst"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00745"},{"key":"ref46","first-page":"1020","article-title":"Structured adaptive and random spinners for fast machine learning computations","author":"bojarski","year":"2017","journal-title":"Proc Artif Intell Statist"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2015.173"},{"key":"ref48","article-title":"Actor-Mimic: Deep multitask and transfer reinforcement learning","author":"parisotto","year":"2016","journal-title":"Proc Int Conf Learn Represent"},{"key":"ref47","first-page":"12024","article-title":"Butterfly transform: An efficient FFT based neural architecture design","author":"vahid","year":"2020","journal-title":"Proc IEEE\/CVF Conf Comput Vis Pattern Recognit (CVPR)"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00291"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00716"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.435"},{"key":"ref43","article-title":"Flattened convolutional neural networks for feedforward acceleration","author":"jin","year":"2015","journal-title":"Proc Int Conf Learn Represent"}],"container-title":["IEEE Access"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/6287639\/9312710\/09446890.pdf?arnumber=9446890","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2021,12,17]],"date-time":"2021-12-17T19:56:26Z","timestamp":1639770986000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/9446890\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021]]},"references-count":73,"URL":"https:\/\/doi.org\/10.1109\/access.2021.3086321","relation":{},"ISSN":["2169-3536"],"issn-type":[{"type":"electronic","value":"2169-3536"}],"subject":[],"published":{"date-parts":[[2021]]}}}