{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,18]],"date-time":"2026-07-18T16:25:51Z","timestamp":1784391951692,"version":"3.55.0"},"reference-count":201,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"2","license":[{"start":{"date-parts":[[2025,2,1]],"date-time":"2025-02-01T00:00:00Z","timestamp":1738368000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc-nd\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100009975","name":"Khalifa University of Science and Technology","doi-asserted-by":"publisher","award":["RIG-2024-066"],"award-info":[{"award-number":["RIG-2024-066"]}],"id":[{"id":"10.13039\/501100009975","id-type":"DOI","asserted-by":"publisher"}]},{"name":"System-on-Chip Center (SoCC) through the Khalifa University Center for Cyber-Physical Systems","award":["RC2-2018-020"],"award-info":[{"award-number":["RC2-2018-020"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Proc. IEEE"],"published-print":{"date-parts":[[2025,2]]},"DOI":"10.1109\/jproc.2025.3578756","type":"journal-article","created":{"date-parts":[[2025,6,26]],"date-time":"2025-06-26T13:43:46Z","timestamp":1750945426000},"page":"172-207","source":"Crossref","is-referenced-by-count":2,"title":["A Survey and Comparative Analysis of Number Systems for Deep Neural Networks"],"prefix":"10.1109","volume":"113","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-7251-747X","authenticated-orcid":false,"given":"Ghada","family":"Alsuhli","sequence":"first","affiliation":[{"name":"Department of Computer and Information Engineering, System on Chip Center, Khalifa University, Abu Dhabi, United Arab Emirates"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2581-3193","authenticated-orcid":false,"given":"Vasilis","family":"Sakellariou","sequence":"additional","affiliation":[{"name":"Department of Computer and Information Engineering, System on Chip Center, Khalifa University, Abu Dhabi, United Arab Emirates"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7185-0278","authenticated-orcid":false,"given":"Hani","family":"Saleh","sequence":"additional","affiliation":[{"name":"Department of Computer and Information Engineering, System on Chip Center, Khalifa University, Abu Dhabi, United Arab Emirates"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9600-8036","authenticated-orcid":false,"given":"Mahmoud","family":"Al-Qutayri","sequence":"additional","affiliation":[{"name":"Department of Computer and Information Engineering, System on Chip Center, Khalifa University, Abu Dhabi, United Arab Emirates"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6063-473X","authenticated-orcid":false,"given":"Baker","family":"Mohammad","sequence":"additional","affiliation":[{"name":"Department of Computer and Information Engineering, System on Chip Center, Khalifa University, Abu Dhabi, United Arab Emirates"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3696-4958","authenticated-orcid":false,"given":"Thanos","family":"Stouraitis","sequence":"additional","affiliation":[{"name":"Department of Computer and Information Engineering, System on Chip Center, Khalifa University, Abu Dhabi, United Arab Emirates"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1016\/j.array.2021.100057"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/TWC.2023.3240969"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.3390\/make3040048"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2020.2979670"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/TCSS.2022.3231701"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/TWC.2023.3327372"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-021-82543-3"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/ISSCC.2019.8662396"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1007\/s10462-020-09816-7"},{"key":"ref10","article-title":"A survey of quantization methods for efficient neural network inference","author":"Gholami","year":"2021","journal-title":"arXiv:2103.13630"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/JPROC.2017.2761740"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/TCSII.2021.3072217"},{"key":"ref13","first-page":"10271","article-title":"Pushing the limits of narrow precision inferencing at cloud scale with Microsoft floating point","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Rouhani"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.2200\/s01004ed1v01y202004cac050"},{"key":"ref15","article-title":"BFloat16: The secret to high performance on cloud TPUs","volume":"30","author":"Wang","year":"2019","journal-title":"Google Cloud Blog"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/MM.2021.3061394"},{"key":"ref17","first-page":"1737","article-title":"Deep learning with limited numerical precision","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Gupta"},{"key":"ref18","article-title":"A study of BFLOAT16 for deep learning training","author":"Kalamkar","year":"2019","journal-title":"arXiv:1905.12322"},{"key":"ref19","first-page":"1742","article-title":"Flexpoint: An adaptive numerical format for efficient training of deep neural networks","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"30","author":"K\u00f6ster"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/ARITH.2018.8464801"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1016\/B978-0-12-820064-3.00004-0"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/ieeestd.2020.9091348"},{"key":"ref23","article-title":"Training deep neural networks with low precision multiplications","author":"Courbariaux","year":"2014","journal-title":"arXiv:1412.7024"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1145\/3448980"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1145\/3474597"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1587\/elex.15.20180909"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/ICEIC49074.2020.9050989"},{"key":"ref28","volume-title":"Tensorfloat-32 in the A100 GPU Accelerates AI Training, HPC Up To 20x","year":"2020"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1109\/ISSCC42613.2021.9365803"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1109\/TCAD.2023.3253045"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1109\/AICAS54282.2022.9869965"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1109\/ICMLA51294.2020.00017"},{"key":"ref33","article-title":"Mixed precision training","author":"Micikevicius","year":"2017","journal-title":"arXiv:1710.03740"},{"key":"ref34","first-page":"4900","article-title":"Hybrid 8-bit floating point (HFP8) training and inference for deep neural networks","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"32","author":"Sun"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1109\/TC.2020.2985971"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.23919\/DATE.2019.8715262"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.14529\/jsfi170206"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2021.3086669"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-37277-4_25"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1109\/CVPRW50498.2020.00371"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1109\/EMC2.2018.00012"},{"key":"ref42","article-title":"Template-based posit multiplication for training and inferring in neural networks","author":"Murillo Montero","year":"2019","journal-title":"arXiv:1907.04091"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1145\/3316279.3316282"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.1109\/CVPRW53098.2021.00346"},{"key":"ref45","article-title":"Study of posit numeric in speech recognition neural inference","author":"Wan","year":"2018"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1109\/ISCAS45731.2020.9180771"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1109\/TETC.2021.3109127"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2021.3098730"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.1016\/j.dsp.2020.102762"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1109\/ISCAS.2019.8702349"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.1109\/SpaceComp.2019.00011"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.1109\/IPDPSW.2018.00029"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.1109\/TVLSI.2021.3131609"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.1007\/s11554-020-00984-x"},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.1109\/SMARTCOMP50058.2020.00053"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.23919\/DATE51398.2021.9473933"},{"key":"ref57","first-page":"519","article-title":"Arbitor: A numerically accurate hardware emulation tool for DNN accelerators","volume-title":"Proc. USENIX Annu. Tech. Conf. (USENIX ATC)","author":"Jiang"},{"key":"ref58","doi-asserted-by":"publisher","DOI":"10.1145\/3458817.3476157"},{"key":"ref59","doi-asserted-by":"publisher","DOI":"10.1109\/ISCA52012.2021.00010"},{"key":"ref60","doi-asserted-by":"publisher","DOI":"10.24963\/ijcai.2020\/520"},{"key":"ref61","doi-asserted-by":"publisher","DOI":"10.1109\/EMC2-NIPS53020.2019.00016"},{"key":"ref62","doi-asserted-by":"publisher","DOI":"10.1109\/MWSCAS48704.2020.9184436"},{"key":"ref63","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2015.7178146"},{"key":"ref64","first-page":"2849","article-title":"Fixed point quantization of deep convolutional networks","volume-title":"Proc. 5th Int. Conf. Learn. Represent. (ICLR)","author":"Lin"},{"key":"ref65","doi-asserted-by":"publisher","DOI":"10.1109\/SiPS.2014.6986082"},{"key":"ref66","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2021.3054879"},{"key":"ref67","article-title":"Mixed low-precision deep learning inference using dynamic fixed point","author":"Mellempudi","year":"2017","journal-title":"arXiv:1701.08978"},{"key":"ref68","doi-asserted-by":"publisher","DOI":"10.1109\/ISMVL52857.2022.00031"},{"key":"ref69","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00066"},{"key":"ref70","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-46493-0_32"},{"key":"ref71","doi-asserted-by":"publisher","DOI":"10.1109\/CVPRW53098.2021.00521"},{"key":"ref72","doi-asserted-by":"publisher","DOI":"10.1109\/JSSC.2019.2963616"},{"key":"ref73","article-title":"Neural networks with few multiplications","author":"Lin","year":"2015","journal-title":"arXiv:1510.03009"},{"key":"ref74","first-page":"55","article-title":"B-DCGAN: Evaluation of binarized DCGAN for FPGA","volume-title":"Proc. Neural Inf. Process.","author":"Hideo"},{"key":"ref75","doi-asserted-by":"publisher","DOI":"10.1109\/TCAD.2021.3075420"},{"key":"ref76","doi-asserted-by":"publisher","DOI":"10.1016\/j.patcog.2020.107281"},{"key":"ref77","doi-asserted-by":"publisher","DOI":"10.1109\/TCSVT.2018.2888898"},{"key":"ref78","doi-asserted-by":"publisher","DOI":"10.1109\/ICPR56361.2022.9956041"},{"key":"ref79","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2021.3129357"},{"key":"ref80","first-page":"27168","article-title":"ZeroQuant: Efficient and affordable post-training quantization for large-scale transformers","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Yao"},{"key":"ref81","doi-asserted-by":"publisher","DOI":"10.1109\/IJCNN54540.2023.10191521"},{"key":"ref82","doi-asserted-by":"publisher","DOI":"10.1109\/WCNC55385.2023.10118781"},{"key":"ref83","doi-asserted-by":"publisher","DOI":"10.1109\/OJCAS.2023.3282929"},{"key":"ref84","doi-asserted-by":"publisher","DOI":"10.23919\/EUSIPCO54536.2021.9615973"},{"key":"ref85","first-page":"1","article-title":"Hardware architecture of graph neural network-enabled motion planner (invited paper)","volume-title":"Proc. 41st IEEE\/ACM Int. Conf. Comput.-Aided Design","author":"Huang"},{"key":"ref86","doi-asserted-by":"publisher","DOI":"10.1109\/ICNN.1997.616150"},{"key":"ref87","doi-asserted-by":"publisher","DOI":"10.1016\/j.compeleceng.2020.106800"},{"key":"ref88","article-title":"Convolutional neural networks using logarithmic data representation","author":"Miyashita","year":"2016","journal-title":"arXiv:1603.01025"},{"key":"ref89","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP40776.2020.9053015"},{"key":"ref90","doi-asserted-by":"publisher","DOI":"10.1145\/3461699"},{"key":"ref91","doi-asserted-by":"publisher","DOI":"10.1109\/MOCAST.2018.8376572"},{"key":"ref92","doi-asserted-by":"publisher","DOI":"10.1109\/TCAD.2018.2857262"},{"key":"ref93","doi-asserted-by":"publisher","DOI":"10.1109\/TEC.1962.5219391"},{"key":"ref94","doi-asserted-by":"publisher","DOI":"10.1016\/j.micpro.2010.07.001"},{"key":"ref95","doi-asserted-by":"publisher","DOI":"10.23919\/DATE.2019.8714868"},{"key":"ref96","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2020.2985345"},{"key":"ref97","doi-asserted-by":"publisher","DOI":"10.1007\/s00034-021-01793-8"},{"key":"ref98","doi-asserted-by":"publisher","DOI":"10.1109\/ASPDAC.2018.8297391"},{"key":"ref99","article-title":"Hardware-efficient transformer training via piecewise affine operations","author":"Kosson","year":"2023","journal-title":"arXiv:2305.17190"},{"key":"ref100","doi-asserted-by":"publisher","DOI":"10.1109\/TC.2018.2880742"},{"key":"ref101","first-page":"145","article-title":"Multiplier-less artificial neurons exploiting error resiliency for energy-efficient neural computing","volume-title":"Proc. IEEE Design, Automat. Test Eur. Conf. Exhib. (DATE)","author":"Sarwar"},{"key":"ref102","doi-asserted-by":"publisher","DOI":"10.1109\/ICCAD.2015.7372600"},{"key":"ref103","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-20282-7_17"},{"key":"ref104","doi-asserted-by":"publisher","DOI":"10.1109\/ARITH.2019.00029"},{"key":"ref105","doi-asserted-by":"publisher","DOI":"10.1109\/TVLSI.2019.2940943"},{"key":"ref106","doi-asserted-by":"publisher","DOI":"10.1109\/TC.2020.2992113"},{"key":"ref107","article-title":"Rethinking floating point for deep learning","author":"Johnson","year":"2018","journal-title":"arXiv:1811.01721"},{"key":"ref108","doi-asserted-by":"publisher","DOI":"10.1109\/ISOCC.2018.8649961"},{"key":"ref109","doi-asserted-by":"publisher","DOI":"10.1080\/02533839.2015.1070692"},{"key":"ref110","doi-asserted-by":"publisher","DOI":"10.1109\/VDAT.2011.5783555"},{"key":"ref111","article-title":"Low-precision training in logarithmic number system using multiplicative weight update","author":"Zhao","year":"2021","journal-title":"arXiv:2106.13914"},{"key":"ref112","doi-asserted-by":"publisher","DOI":"10.1145\/3240765.3240803"},{"key":"ref113","article-title":"A very compact embedded CNN processor design based on logarithmic computing","author":"Lu","year":"2020","journal-title":"arXiv:2010.11686"},{"key":"ref114","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2017.7953288"},{"key":"ref115","doi-asserted-by":"publisher","DOI":"10.1109\/SOCC.2018.8618560"},{"key":"ref116","doi-asserted-by":"publisher","DOI":"10.1007\/s11265-020-01557-8"},{"key":"ref117","doi-asserted-by":"publisher","DOI":"10.14569\/IJACSA.2018.091062"},{"key":"ref118","doi-asserted-by":"publisher","DOI":"10.1109\/IIAI-AAI.2018.00133"},{"key":"ref119","volume-title":"Residue Number System Arithmetic: Modern Applications in Digital Signal Processing","author":"Soderstrand","year":"1986"},{"key":"ref120","article-title":"Residue number systems: A survey","author":"Nannarelli","year":"2008"},{"key":"ref121","doi-asserted-by":"publisher","DOI":"10.1109\/TC.2002.1146705"},{"key":"ref122","doi-asserted-by":"publisher","DOI":"10.1109\/82.958337"},{"key":"ref123","doi-asserted-by":"publisher","DOI":"10.1007\/s00200-015-0268-1"},{"key":"ref124","doi-asserted-by":"publisher","DOI":"10.1109\/TC.2005.63"},{"key":"ref125","doi-asserted-by":"publisher","DOI":"10.1109\/ISCAS.2009.5117800"},{"key":"ref126","doi-asserted-by":"publisher","DOI":"10.1109\/TVLSI.2015.2484618"},{"key":"ref127","doi-asserted-by":"publisher","DOI":"10.1109\/APSIPA.2016.7820790"},{"issue":"2","key":"ref128","doi-asserted-by":"crossref","first-page":"379","DOI":"10.1109\/TVLSI.2014.2308014","article-title":"Fast sign detection algorithm for the RNS moduli set {2n+1-1,2n-1,2n}","volume":"23","author":"Xu","year":"2015","journal-title":"IEEE Trans. Very Large Scale Integr. (VLSI) Syst."},{"key":"ref129","doi-asserted-by":"publisher","DOI":"10.1109\/TVLSI.2003.810799"},{"key":"ref130","doi-asserted-by":"publisher","DOI":"10.1109\/TVLSI.2008.2004550"},{"key":"ref131","doi-asserted-by":"publisher","DOI":"10.1109\/ISCAS.2018.8351352"},{"key":"ref132","doi-asserted-by":"publisher","DOI":"10.1109\/FPL.2015.7293933"},{"key":"ref133","doi-asserted-by":"crossref","first-page":"232","DOI":"10.1016\/j.matcom.2020.04.031","article-title":"Application of the residue number system to reduce hardware costs of the convolutional neural network implementation","volume":"177","author":"Valueva","year":"2020","journal-title":"Math. Comput. Simul."},{"key":"ref134","article-title":"Applying the residue number system to network inference","author":"Abdelhamid","year":"2017","journal-title":"arXiv:1712.04614"},{"key":"ref135","first-page":"222","article-title":"Residue-net: Multiplication-free neural network by in-situ no-loss migration to residue number systems","volume-title":"Proc. 26th Asia South Pacific Design Autom. Conf. (ASP-DAC)","author":"Salamat"},{"key":"ref136","doi-asserted-by":"publisher","DOI":"10.1109\/TCSI.2019.2951083"},{"key":"ref137","volume-title":"Residue Arithmetic and Its Applications to Computer Technology","author":"Szab\u00f3","year":"1967"},{"key":"ref138","doi-asserted-by":"publisher","DOI":"10.1145\/3007787.3001177"},{"key":"ref139","doi-asserted-by":"publisher","DOI":"10.1109\/TETC.2023.3301590"},{"key":"ref140","doi-asserted-by":"publisher","DOI":"10.1109\/ICRC.2018.8638592"},{"key":"ref141","doi-asserted-by":"publisher","DOI":"10.1109\/ICCAD51958.2021.9643531"},{"key":"ref142","doi-asserted-by":"publisher","DOI":"10.1109\/ISCAS48785.2022.9937633"},{"key":"ref143","doi-asserted-by":"publisher","DOI":"10.1109\/AICAS59952.2024.10595885"},{"key":"ref144","doi-asserted-by":"publisher","DOI":"10.1109\/TVLSI.2019.2913958"},{"key":"ref145","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v32i1.11334"},{"key":"ref146","first-page":"7015","article-title":"SWALP: Stochastic weight averaging in low precision training","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Yang"},{"key":"ref147","doi-asserted-by":"publisher","DOI":"10.1109\/ICFPT47387.2019.00012"},{"key":"ref148","doi-asserted-by":"publisher","DOI":"10.1109\/APCCAS50809.2020.9301687"},{"key":"ref149","doi-asserted-by":"publisher","DOI":"10.1109\/GlobalSIP45357.2019.8969292"},{"key":"ref150","doi-asserted-by":"publisher","DOI":"10.1109\/FPL.2018.00056"},{"key":"ref151","article-title":"FAST: DNN training under variable precision block floating point with stochastic rounding","author":"Zhang","year":"2021","journal-title":"arXiv:2110.15456"},{"key":"ref152","first-page":"451","article-title":"Training DNNs with hybrid block floating point","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"31","author":"Drumond"},{"key":"ref153","doi-asserted-by":"publisher","DOI":"10.1109\/ISVLSI51109.2021.00048"},{"key":"ref154","doi-asserted-by":"publisher","DOI":"10.1145\/3020078.3021738"},{"key":"ref155","doi-asserted-by":"publisher","DOI":"10.1109\/ICCD.2018.00067"},{"key":"ref156","article-title":"Mixed precision training of convolutional neural networks using integer operations","author":"Das","year":"2018","journal-title":"arXiv:1802.00930"},{"key":"ref157","first-page":"82","article-title":"Processing element architecture design for deep reinforcement learning with flexible block floating point exploiting signal statistics","volume-title":"Proc. Asia\u2013Pacific Signal Inf. Process. Assoc. Annu. Summit Conf. (APSIPA ASC)","author":"Su"},{"key":"ref158","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2021.3116302"},{"key":"ref159","first-page":"2038","article-title":"Model fusion with Kullback\u2013Leibler divergence","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Claici"},{"key":"ref160","article-title":"FlexBlock: A flexible DNN training accelerator with multi-mode block floating point support","author":"Noh","year":"2022","journal-title":"arXiv:2203.06673"},{"key":"ref161","doi-asserted-by":"publisher","DOI":"10.1109\/ICFPT47387.2019.00009"},{"key":"ref162","first-page":"1","article-title":"A block minifloat representation for training deep neural networks","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Fox"},{"key":"ref163","doi-asserted-by":"publisher","DOI":"10.1109\/ISCMI51676.2020.9311563"},{"key":"ref164","doi-asserted-by":"publisher","DOI":"10.1109\/VLSI-DAT.2019.8742040"},{"key":"ref165","doi-asserted-by":"publisher","DOI":"10.1145\/3203217.3203282"},{"key":"ref166","doi-asserted-by":"publisher","DOI":"10.1109\/ISPACS48206.2019.8986370"},{"key":"ref167","doi-asserted-by":"publisher","DOI":"10.1109\/JETCAS.2020.3015753"},{"key":"ref168","article-title":"Hardware-oriented approximation of convolutional neural networks","author":"Gysel","year":"2016","journal-title":"arXiv:1604.03168"},{"key":"ref169","doi-asserted-by":"publisher","DOI":"10.1145\/2934583.2934625"},{"key":"ref170","doi-asserted-by":"publisher","DOI":"10.1007\/978-981-10-3159-5_10"},{"key":"ref171","doi-asserted-by":"publisher","DOI":"10.1109\/CAC.2017.8243585"},{"key":"ref172","article-title":"Deep convolutional neural network inference with floating-point weights and fixed-point activations","author":"Lai","year":"2017","journal-title":"arXiv:1703.03073"},{"key":"ref173","doi-asserted-by":"publisher","DOI":"10.1109\/CoolChips.2017.7946376"},{"key":"ref174","doi-asserted-by":"publisher","DOI":"10.1109\/TCSI.2018.2880363"},{"key":"ref175","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2022.3157893"},{"key":"ref176","doi-asserted-by":"publisher","DOI":"10.1109\/ISOCC53507.2021.9614019"},{"key":"ref177","doi-asserted-by":"publisher","DOI":"10.1109\/FPL50879.2020.00056"},{"key":"ref178","article-title":"Implementation of an 8-bit dynamic fixed-point convolutional neural network for human sign language recognition on a Xilinx FPGA board","author":"Prieto","year":"2019"},{"key":"ref179","doi-asserted-by":"publisher","DOI":"10.1109\/EDSSC.2019.8754067"},{"key":"ref180","doi-asserted-by":"publisher","DOI":"10.1109\/ICIP.2019.8803490"},{"key":"ref181","doi-asserted-by":"publisher","DOI":"10.1109\/IJCNN.2017.7966324"},{"key":"ref182","article-title":"Quantization error as a metric for dynamic precision scaling in neural net training","author":"Taras","year":"2018","journal-title":"arXiv:1801.08621"},{"key":"ref183","doi-asserted-by":"publisher","DOI":"10.1109\/JSSC.2021.3066400"},{"key":"ref184","first-page":"14651","article-title":"FP8 quantization: The power of the exponent","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Kuzmin"},{"key":"ref185","article-title":"QLoRA: Efficient finetuning of quantized LLMs","author":"Dettmers","year":"2023","journal-title":"arXiv:2305.14314"},{"key":"ref186","article-title":"8-bit optimizers via block-wise quantization","author":"Dettmers","year":"2021","journal-title":"arXiv:2110.02861"},{"key":"ref187","article-title":"OPTQ: Accurate quantization for generative pre-trained transformers","volume-title":"Proc. 11th Int. Conf. Learn. Represent.","author":"Frantar"},{"key":"ref188","doi-asserted-by":"publisher","DOI":"10.1109\/MICRO50266.2020.00071"},{"key":"ref189","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref190","first-page":"2383","article-title":"SQuAD: 100,000+ questions for machine comprehension of text","volume-title":"Proc. Conf. Empirical Methods Natural Lang. Process.","author":"Rajpurkar"},{"key":"ref191","first-page":"4475","article-title":"Optimal brain compression: A framework for accurate post-training quantization and pruning","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Frantar"},{"key":"ref192","first-page":"1","article-title":"Learning both weights and connections for efficient neural network","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"28","author":"Han"},{"key":"ref193","doi-asserted-by":"publisher","DOI":"10.1109\/ISSCC.2014.6757323"},{"key":"ref194","first-page":"223","article-title":"Rethinking floating point overheads for mixed precision DNN accelerators","volume":"3","author":"Abdel-Aziz","year":"2021","journal-title":"Proc. Mach. Learn. Syst."},{"key":"ref195","doi-asserted-by":"publisher","DOI":"10.11591\/ijeecs.v19.i3.pp1480-1489"},{"key":"ref196","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.00240"},{"key":"ref197","doi-asserted-by":"publisher","DOI":"10.1109\/TC.2019.2930514"},{"key":"ref198","article-title":"Overcoming challenges in fixed point training of deep convolutional networks","author":"Lin","year":"2016","journal-title":"arXiv:1607.02241"},{"key":"ref199","doi-asserted-by":"publisher","DOI":"10.1109\/IJCNN.2017.7966159"},{"key":"ref200","doi-asserted-by":"publisher","DOI":"10.1109\/DAC18072.2020.9218516"},{"key":"ref201","first-page":"49146","article-title":"Training transformers with 4-bit integers","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Xi"}],"container-title":["Proceedings of the IEEE"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/5\/11082589\/11053145.pdf?arnumber=11053145","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,7,18]],"date-time":"2025-07-18T04:49:49Z","timestamp":1752814189000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11053145\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,2]]},"references-count":201,"journal-issue":{"issue":"2"},"URL":"https:\/\/doi.org\/10.1109\/jproc.2025.3578756","relation":{},"ISSN":["0018-9219","1558-2256"],"issn-type":[{"value":"0018-9219","type":"print"},{"value":"1558-2256","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,2]]}}}