{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,2,21]],"date-time":"2025-02-21T10:18:16Z","timestamp":1740133096409,"version":"3.37.3"},"reference-count":54,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"2","funder":[{"DOI":"10.13039\/501100001459","name":"Ministry of Education - Singapore","doi-asserted-by":"publisher","award":["T1-251RES1818","MOE2016-T2-2-150"],"award-info":[{"award-number":["T1-251RES1818","MOE2016-T2-2-150"]}],"id":[{"id":"10.13039\/501100001459","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Parallel Distrib. Syst."],"published-print":{"date-parts":[[2022,2,1]]},"DOI":"10.1109\/tpds.2021.3093239","type":"journal-article","created":{"date-parts":[[2021,6,29]],"date-time":"2021-06-29T19:33:27Z","timestamp":1624995207000},"page":"429-443","source":"Crossref","is-referenced-by-count":5,"title":["Tensorox: Accelerating GPU Applications via Neural Approximation on Unused Tensor Cores"],"prefix":"10.1109","volume":"33","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-3864-8027","authenticated-orcid":false,"given":"Nhut-Minh","family":"Ho","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4281-2053","authenticated-orcid":false,"given":"Weng-Fai","family":"Wong","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00286"},{"key":"ref38","first-page":"5145","article-title":"Scalable methods for 8-bit training of neural networks","author":"banner","year":"2018","journal-title":"Proc Neural Inf Process Syst"},{"key":"ref33","article-title":"GPU performance analysis and optimization","author":"micikevicius","year":"2012","journal-title":"Proc GPU Technol Conf"},{"article-title":"Dissecting the Nvidia turing T4 GPU via microbenchmarking","year":"2019","author":"jia","key":"ref32"},{"article-title":"Dissecting the Nvidia volta GPU architecture via microbenchmarking","year":"2018","author":"jia","key":"ref31"},{"article-title":"Adam: A method for stochastic optimization","year":"2014","author":"kingma","key":"ref30"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1109\/HPCA.2016.7446063"},{"key":"ref36","first-page":"1","article-title":"Canny edge detection on Nvidia CUDA","author":"luo","year":"2008","journal-title":"Proc IEEE Comput Soc Conf Comput Vis Pattern Recognit Workshops"},{"year":"2020","key":"ref35","article-title":"CUDA SDK samples"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1016\/S0167-8655(00)00112-4"},{"article-title":"Automatic differentiation in pytorch","year":"0","author":"paszke","key":"ref28"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/MDAT.2016.2630270"},{"key":"ref29","first-page":"2825","article-title":"Scikit-learn: Machine learning in Python","volume":"12","author":"pedregosa","year":"2011","journal-title":"J Mach Learn Res"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1145\/3079856.3080246"},{"year":"2020","key":"ref1","article-title":"Nvidia tesla p100 whitepaper."},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1145\/2503210.2503296"},{"year":"2020","key":"ref22","article-title":"CUDA wmma instructions."},{"key":"ref21","first-page":"1","article-title":"The comparative power of relu networks and polynomial kernels in the presence of sparse latent structure","author":"koehler","year":"2019","journal-title":"Proc Int Conf Learn Representations"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1145\/2540708.2540711"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1145\/2644865.2541948"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1016\/0925-2312(94)E0053-T"},{"key":"ref25","first-page":"1","article-title":"Multi-scale context aggregation by dilated convolutions","author":"yu","year":"2016","journal-title":"Proc 4th Int Conf Learn Representations"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.23919\/DATE.2018.8342187"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.1145\/3218603.3218626"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.155"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.1145\/3005348"},{"key":"ref52","first-page":"1737","article-title":"Deep learning with limited numerical precision","author":"gupta","year":"2015","journal-title":"Proc Int Conf Mach Learn"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijpe.2019.02.017"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1016\/j.renene.2018.11.061"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1145\/2893356"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1016\/j.compeleceng.2017.09.016"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"article-title":"Attention is all you need","year":"2017","author":"vaswani","key":"ref14"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-019-1923-7"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1016\/j.ymssp.2020.106625"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/MICRO.2012.48"},{"key":"ref18","doi-asserted-by":"crossref","DOI":"10.1038\/nature14539","article-title":"Deep learning","volume":"521","author":"lecun","year":"2015","journal-title":"Nature"},{"article-title":"Resolving numerical anomalies in scientific computation","year":"0","author":"bailey","key":"ref19"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1109\/HPEC.2017.8091072"},{"year":"2020","key":"ref3","article-title":"Nvidia tesla v100 GPU achitecture"},{"year":"2020","key":"ref6","article-title":"Google cloud TPU"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1145\/2830772.2830810"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/SC.2018.00050"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1145\/3330345.3331057"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.1145\/3061639.3062210"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/TII.2017.2786782"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-23702-7_24"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1145\/2025113.2025133"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1145\/3307650.3322215"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1145\/3316781.3317776"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1109\/LES.2017.2758679"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1109\/MDAT.2015.2505723"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.1145\/3330345.3330360"},{"key":"ref43","first-page":"443","author":"wong","year":"2019","journal-title":"Compilation and Other Software Techniques Enabling Approximate Computing Methodologies and CAD"}],"container-title":["IEEE Transactions on Parallel and Distributed Systems"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/71\/9493664\/09468379.pdf?arnumber=9468379","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2021,11,8]],"date-time":"2021-11-08T23:00:16Z","timestamp":1636412416000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/9468379\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,2,1]]},"references-count":54,"journal-issue":{"issue":"2"},"URL":"https:\/\/doi.org\/10.1109\/tpds.2021.3093239","relation":{},"ISSN":["1045-9219","1558-2183","2161-9883"],"issn-type":[{"type":"print","value":"1045-9219"},{"type":"electronic","value":"1558-2183"},{"type":"electronic","value":"2161-9883"}],"subject":[],"published":{"date-parts":[[2022,2,1]]}}}