{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,6,10]],"date-time":"2025-06-10T18:10:10Z","timestamp":1749579010888,"version":"3.41.0"},"reference-count":71,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","license":[{"start":{"date-parts":[[2025,1,1]],"date-time":"2025-01-01T00:00:00Z","timestamp":1735689600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/legalcode"}],"funder":[{"DOI":"10.13039\/501100002241","name":"Japan Science and Technology Agency","doi-asserted-by":"publisher","award":["JPMJAN24F3"],"award-info":[{"award-number":["JPMJAN24F3"]}],"id":[{"id":"10.13039\/501100002241","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Access"],"published-print":{"date-parts":[[2025]]},"DOI":"10.1109\/access.2025.3568916","type":"journal-article","created":{"date-parts":[[2025,5,12]],"date-time":"2025-05-12T17:45:14Z","timestamp":1747071914000},"page":"86509-86527","source":"Crossref","is-referenced-by-count":0,"title":["<i>WhiteDwarf<\/i>: A Holistic Co-Design Approach to Ultra-Compact Neural Inference Acceleration"],"prefix":"10.1109","volume":"13","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-3206-1479","authenticated-orcid":false,"given":"\u00c1ngel L\u00f3pez","family":"Garc\u00eda-Arias","sequence":"first","affiliation":[{"name":"AI Computing Research Unit, Institute of Science Tokyo (formerly Tokyo Institute of Technology), Yokohama, Kanagawa, Japan"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0005-8472-7841","authenticated-orcid":false,"given":"Yasuyuki","family":"Okoshi","sequence":"additional","affiliation":[{"name":"AI Computing Research Unit, Institute of Science Tokyo (formerly Tokyo Institute of Technology), Yokohama, Kanagawa, Japan"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6639-7694","authenticated-orcid":false,"given":"Jaehoon","family":"Yu","sequence":"additional","affiliation":[{"name":"AI Computing Research Unit, Institute of Science Tokyo (formerly Tokyo Institute of Technology), Yokohama, Kanagawa, Japan"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0002-5273-8448","authenticated-orcid":false,"given":"Junnosuke","family":"Suzuki","sequence":"additional","affiliation":[{"name":"AI Computing Research Unit, Institute of Science Tokyo (formerly Tokyo Institute of Technology), Yokohama, Kanagawa, Japan"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0004-8418-4691","authenticated-orcid":false,"given":"Hikari","family":"Otsuka","sequence":"additional","affiliation":[{"name":"AI Computing Research Unit, Institute of Science Tokyo (formerly Tokyo Institute of Technology), Yokohama, Kanagawa, Japan"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0795-2974","authenticated-orcid":false,"given":"Kazushi","family":"Kawamura","sequence":"additional","affiliation":[{"name":"AI Computing Research Unit, Institute of Science Tokyo (formerly Tokyo Institute of Technology), Yokohama, Kanagawa, Japan"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2003-2574","authenticated-orcid":false,"given":"Thiem","family":"Van Chu","sequence":"additional","affiliation":[{"name":"AI Computing Research Unit, Institute of Science Tokyo (formerly Tokyo Institute of Technology), Yokohama, Kanagawa, Japan"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7949-0417","authenticated-orcid":false,"given":"Daichi","family":"Fujiki","sequence":"additional","affiliation":[{"name":"AI Computing Research Unit, Institute of Science Tokyo (formerly Tokyo Institute of Technology), Yokohama, Kanagawa, Japan"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1543-1252","authenticated-orcid":false,"given":"Masato","family":"Motomura","sequence":"additional","affiliation":[{"name":"AI Computing Research Unit, Institute of Science Tokyo (formerly Tokyo Institute of Technology), Yokohama, Kanagawa, Japan"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","first-page":"382","article-title":"Bit-pragmatic deep neural network computing","volume-title":"Proc. 50th Annu. IEEE\/ACM Int. Symp. Microarchit. (MICRO)","author":"Albericio"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/ISCA.2016.11"},{"key":"ref3","first-page":"242","article-title":"A convergence theory for deep learning via over-parameterization","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Allen-Zhu"},{"key":"ref4","article-title":"Estimating or propagating gradients through stochastic neurons for conditional computation","author":"Bengio","year":"2013","journal-title":"arXiv:1308.3432"},{"key":"ref5","first-page":"1","article-title":"What is the state of neural network pruning","volume-title":"Proc. Mach. Learn. Syst.","author":"Blalock"},{"article-title":"Peek-a-Boo: What (more) is disguised in a randomly weighted neural network, and how to find it efficiently","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Chen","key":"ref6"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/ISSCC.2016.7418007"},{"key":"ref8","first-page":"4503","article-title":"Pruning randomly initialized neural networks with iterative randomization","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Chijiwa"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/CVPRW50498.2020.00359"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1145\/3297858.3304041"},{"article-title":"An image is worth 16\u00d716 words: Transformers for image recognition at scale","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Dosovitskiy","key":"ref11"},{"key":"ref12","first-page":"6910","article-title":"Learned step size quantization","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Esser"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.l803.03635"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.23919\/VLSICircuits52068.2021.9492338"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1145\/3352460.3358291"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/ISCA.2016.30"},{"article-title":"Deep compression: Compressing deep neural networks with pruning, trained quantization and Huffman coding","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Han","key":"ref17"},{"key":"ref18","first-page":"1135","article-title":"Learning both weights and connections for efficient neural network","volume-title":"Proc. NIPS","author":"Han"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/ISSCC42614.2022.9731668"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/ISSCC.2014.6757323"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2022.3145427"},{"key":"ref23","article-title":"MobileNets: Efficient convolutional neural networks for mobile vision applications","author":"Howard","year":"2017","journal-title":"arXiv:1704.04861"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1109\/JETCAS.2019.2910232"},{"key":"ref25","article-title":"Residual connections encourage iterative inference","author":"Jastrz\u0229bski","year":"2017","journal-title":"arXiv:1710.04773"},{"article-title":"Signing the supermask: Keep, hide, invert","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Koster","key":"ref26"},{"volume-title":"Learning multiple layers of features from tiny images","year":"2009","author":"Krizhevsky","key":"ref27"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1145\/3065386"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1038\/nature14539"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1109\/MICRO56248.2022.00097"},{"key":"ref31","article-title":"Bridging the gaps between residual learning, recurrent neural networks and visual cortex","author":"Liao","year":"2016","journal-title":"arXiv:1604.03640"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1109\/DAC56929.2023.10247662"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1109\/HPCA53966.2022.00049"},{"article-title":"Hidden-fold networks: Random recurrent residuals using sparse supermasks","volume-title":"Proc. British Mach. Vis. Conf.","author":"L\u00f3pez Garc\u00eda-Arias","key":"ref34"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2023.3245808"},{"key":"ref36","article-title":"MDMLP: Image classification from scratch on small datasets with MLP","author":"Lv","year":"2022","journal-title":"arXiv:2205.14477"},{"key":"ref37","article-title":"FP8 formats for deep learning","author":"Micikevicius","year":"2022","journal-title":"arXiv:2209.05433"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/ISSCC42615.2023.10067615"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.23919\/VLSITechnologyandCir57934.2023.10185264"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1109\/isscc49657.2024.10454357"},{"key":"ref41","first-page":"17045","article-title":"Multicoated supermasks enhance hidden networks","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Okoshi"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1145\/3079856.3080254"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1109\/VLSITechnologyandCir46783.2024.10631515"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.1016\/j.patcog.2020.107281"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.01191"},{"key":"ref46","first-page":"8821","article-title":"Zero-shot text-to-image generation","volume-title":"Proc. Int. Conf. Mach. Learn. (ICML)","author":"Ramesh"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-015-0816-y"},{"key":"ref48","first-page":"304","article-title":"Laconic deep learning inference acceleration","volume-title":"Proc. ACM\/IEEE 46th Annu. Int. Symp. Comput. Archit. (ISCA)","author":"Sharify"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/P19-1355"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1109\/ISSCC42615.2023.10067644"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-01766-7"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.1109\/ISSCC42615.2023.10067817"},{"key":"ref53","first-page":"24261","article-title":"MLP-Mixer: An all-MLP architecture for vision","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"34","author":"Tolstikhin"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2022.3206148"},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00821"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2018.2886192"},{"key":"ref57","doi-asserted-by":"publisher","DOI":"10.48550\/ARXIV.1706.03762"},{"key":"ref58","doi-asserted-by":"publisher","DOI":"10.1109\/ISSCC42614.2022.9731686"},{"key":"ref59","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.00499"},{"key":"ref60","article-title":"Multicoated and folded graph neural networks with strong lottery tickets","volume-title":"Proc. 2nd Learn. Graphs Conf.","volume":"231","author":"Yan"},{"article-title":"BSQ: Exploring bit-level sparsity for mixed-precision neural network quantization","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Yang","key":"ref61"},{"key":"ref62","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00612"},{"key":"ref63","doi-asserted-by":"publisher","DOI":"10.5244\/C.30.87"},{"key":"ref64","doi-asserted-by":"publisher","DOI":"10.1109\/JSSC.2020.3043870"},{"key":"ref65","doi-asserted-by":"publisher","DOI":"10.1109\/MICRO.2016.7783723"},{"key":"ref66","doi-asserted-by":"publisher","DOI":"10.1109\/EMC2-NIPS53020.2019.00010"},{"key":"ref67","article-title":"Integer or floating point? New outlooks for low-bit quantization on large language models","author":"Zhang","year":"2023","journal-title":"arXiv:2305.12356"},{"key":"ref68","article-title":"Learning N: M fine-grained structured sparse neural networks from scratch","author":"Zhou","year":"2021","journal-title":"arXiv:2102.04010"},{"key":"ref69","first-page":"3597","article-title":"Deconstructing lottery tickets: Zeros, signs, and the supermask","volume-title":"Proc. NIPS","author":"Zhou"},{"key":"ref70","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.00360"},{"key":"ref71","article-title":"To prune, or not to prune: Exploring the efficacy of pruning for model compression","author":"Zhu","year":"2017","journal-title":"arXiv:1710.01878"}],"container-title":["IEEE Access"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/6287639\/10820123\/11000117.pdf?arnumber=11000117","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,10]],"date-time":"2025-06-10T17:53:26Z","timestamp":1749578006000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11000117\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025]]},"references-count":71,"URL":"https:\/\/doi.org\/10.1109\/access.2025.3568916","relation":{},"ISSN":["2169-3536"],"issn-type":[{"type":"electronic","value":"2169-3536"}],"subject":[],"published":{"date-parts":[[2025]]}}}