{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,14]],"date-time":"2026-04-14T16:11:20Z","timestamp":1776183080041,"version":"3.50.1"},"reference-count":31,"publisher":"Institute of Electronics, Information and Communications Engineers (IEICE)","issue":"16","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEICE Electron. Express"],"published-print":{"date-parts":[[2022,8,25]]},"DOI":"10.1587\/elex.19.20220229","type":"journal-article","created":{"date-parts":[[2022,7,12]],"date-time":"2022-07-12T22:13:23Z","timestamp":1657664003000},"page":"20220229-20220229","source":"Crossref","is-referenced-by-count":1,"title":["Sample-wise dynamic precision quantization for neural network acceleration"],"prefix":"10.1587","volume":"19","author":[{"given":"Bowen","family":"Li","sequence":"first","affiliation":[{"name":"School of Micro-Nano Electronics, Zhejiang University"}]},{"given":"Dongliang","family":"Xiong","sequence":"additional","affiliation":[{"name":"School of Micro-Nano Electronics, Zhejiang University"}]},{"given":"Kai","family":"Huang","sequence":"additional","affiliation":[{"name":"School of Micro-Nano Electronics, Zhejiang University"}]},{"given":"Xiaowen","family":"Jiang","sequence":"additional","affiliation":[{"name":"School of Micro-Nano Electronics, Zhejiang University"}]},{"given":"Hao","family":"Yao","sequence":"additional","affiliation":[{"name":"Digital Grid Research Institute, China Southern Power Grid"}]},{"given":"Junjian","family":"Chen","sequence":"additional","affiliation":[{"name":"Digital Grid Research Institute, China Southern Power Grid"}]},{"given":"Luc","family":"Claesen","sequence":"additional","affiliation":[{"name":"Engineering Technology-Electronics-ICT Department, University of Hasselt"}]}],"member":"532","reference":[{"key":"1","doi-asserted-by":"crossref","unstructured":"[1] K. He, <i>et al.<\/i>: \u201cDeep residual learning for image recognition,\u201d IEEE Conference on Computer Vision and Pattern Recognition, CVPR (2016) 770 (DOI: 10.1109\/CVPR.2016.90).","DOI":"10.1109\/CVPR.2016.90"},{"key":"2","doi-asserted-by":"crossref","unstructured":"[2] R.B. Girshick, <i>et al.<\/i>: \u201cRich feature hierarchies for accurate object detection and semantic segmentation,\u201d CVPR (2014) 580 (DOI: 10.1109\/CVPR.2014.81).","DOI":"10.1109\/CVPR.2014.81"},{"key":"3","doi-asserted-by":"crossref","unstructured":"[3] Y. Chen, <i>et al.<\/i>: \u201cEyeriss: an energy-efficient reconfigurable accelerator for deep convolutional neural networks,\u201d IEEE J. Solid State Circuits <b>52<\/b> (2017) 127 (DOI: 10.1109\/JSSC.2016.2616357).","DOI":"10.1109\/JSSC.2016.2616357"},{"key":"4","doi-asserted-by":"crossref","unstructured":"[4] T. Luo, <i>et al.<\/i>: \u201cDaDianNao: a neural network supercomputer,\u201d IEEE Trans. Comput. <b>66<\/b> (2017) 73 (DOI: 10.1109\/TC.2016.2574353).","DOI":"10.1109\/TC.2016.2574353"},{"key":"5","doi-asserted-by":"crossref","unstructured":"[5] S. Scardapane, <i>et al.<\/i>: \u201cWhy should we add early exits to neural networks?,\u201d Cogn. Comput. <b>12<\/b> (2020) 954 (DOI: 10.1007\/s12559-020-09734-4).","DOI":"10.1007\/s12559-020-09734-4"},{"key":"6","unstructured":"[6] J. Yu, <i>et al.<\/i>: \u201cSlimmable neural networks,\u201d ICLR (2019)."},{"key":"7","unstructured":"[7] H. Cai, <i>et al.<\/i>: \u201cOnce-for-all: train one network and specialize it for efficient deployment,\u201d ICLR (2020)."},{"key":"8","doi-asserted-by":"crossref","unstructured":"[8] V. Akhlaghi, <i>et al.<\/i>: \u201cSnaPEA: predictive early activation for reducing computation in deep convolutional neural networks,\u201d ISCA (2018) 662 (DOI: 10.1109\/ISCA.2018.00061).","DOI":"10.1109\/ISCA.2018.00061"},{"key":"9","doi-asserted-by":"crossref","unstructured":"[9] M. Song, <i>et al.<\/i>: \u201cPrediction based execution on deep neural networks,\u201d ISCA (2018) 752 (DOI: 10.1109\/ISCA.2018.00068).","DOI":"10.1109\/ISCA.2018.00068"},{"key":"10","unstructured":"[10] N.P. Jouppi, <i>et al.<\/i>: \u201cIn-datacenter performance analysis of a tensor processing unit,\u201d ISCA (2017) 1 (DOI: 10.1145\/3079856.3080246)."},{"key":"11","doi-asserted-by":"crossref","unstructured":"[11] J. Wu, <i>et al.<\/i>: \u201cQuantized convolutional neural networks for mobile devices,\u201d CVPR (2016) 4820 (DOI: 10.1109\/CVPR.2016.521).","DOI":"10.1109\/CVPR.2016.521"},{"key":"12","doi-asserted-by":"crossref","unstructured":"[12] F. Zhang, <i>et al.<\/i>: \u201cHFOD: a hardware-friendly quantization method for object detection on embedded FPGAs,\u201d IEICE Electron. Express <b>19<\/b> (2022) 20220067 (DOI: 10.1587\/elex.19.20220067).","DOI":"10.1587\/elex.19.20220067"},{"key":"13","doi-asserted-by":"crossref","unstructured":"[13] Q. Jin, <i>et al.<\/i>: \u201cAdaBits: neural network quantization with adaptive bit-widths,\u201d IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR) (2020) 2143 (DOI: 10.1109\/CVPR42600.2020.00222).","DOI":"10.1109\/CVPR42600.2020.00222"},{"key":"14","doi-asserted-by":"crossref","unstructured":"[14] A. Bulat and G. Tzimiropoulos: \u201cBit-mixer: mixed-precision networks with runtime bit-width selection,\u201d ICCV (2021) 5168 (DOI: 10.1109\/ICCV48922.2021.00514).","DOI":"10.1109\/ICCV48922.2021.00514"},{"key":"15","doi-asserted-by":"crossref","unstructured":"[15] H. Yu, <i>et al.<\/i>: \u201cAny-Precision Deep Neural Networks,\u201d AAAI (2021) 10763.","DOI":"10.37686\/ejai.v1i1.82"},{"key":"16","doi-asserted-by":"crossref","unstructured":"[16] H. Sharma, <i>et al.<\/i>: \u201cBit fusion: bit-level dynamically composable architecture for accelerating deep neural network,\u201d 45th ACM\/IEEE Annual International Symposium on Computer Architecture (ISCA) (2018) 764 (DOI: 10.1109\/ISCA.2018.00069).","DOI":"10.1109\/ISCA.2018.00069"},{"key":"17","doi-asserted-by":"crossref","unstructured":"[17] P. Judd, <i>et al.<\/i>: \u201cStripes: bit-serial deep neural network computing,\u201d 49th Annual IEEE\/ACM International Symposium on Microarchitecture (MICRO) (2016) 19: 1 (DOI: 10.1109\/MICRO.2016.7783722).","DOI":"10.1109\/MICRO.2016.7783722"},{"key":"18","doi-asserted-by":"crossref","unstructured":"[18] J. Lee, <i>et al.<\/i>: \u201cUNPU: a 50.6TOPS\/W unified deep neural network accelerator with 1b-to-16b fully-variable weight bit-precision,\u201d ISSCC (2018) 218 (DOI: 10.1109\/ISSCC.2018.8310262).","DOI":"10.1109\/ISSCC.2018.8310262"},{"key":"19","doi-asserted-by":"crossref","unstructured":"[19] S. Zhang, <i>et al.<\/i>: \u201cThread: towards fine-grained precision reconfiguration in variable-precision neural network accelerator,\u201d IEICE Electron. Express <b>16<\/b> (2019) 20190145 (DOI: 10.1587\/elex.16.20190145).","DOI":"10.1587\/elex.16.20190145"},{"key":"20","doi-asserted-by":"crossref","unstructured":"[20] Z. Chen, <i>et al.<\/i>: \u201cYou look twice: GaterNet for dynamic filter selection in CNNs,\u201d IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (2019) 9172 (DOI: 10.1109\/CVPR.2019.00939).","DOI":"10.1109\/CVPR.2019.00939"},{"key":"21","doi-asserted-by":"crossref","unstructured":"[21] C. Li, <i>et al.<\/i>: \u201cDynamic slimmable network,\u201d IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (2021) 8607 (DOI: 10.1109\/cvpr46437.2021.00850).","DOI":"10.1109\/CVPR46437.2021.00850"},{"key":"22","unstructured":"[22] B. Li, <i>et al.<\/i>: \u201cDPOQ: dynamic precision onion quantization,\u201d Asian Conference on Machine Learning (ACML) (2021) 502."},{"key":"23","unstructured":"[23] Y. Kaya, <i>et al.<\/i>: \u201cShallow-deep networks: understanding and mitigating network overthinking,\u201d Proceedings of the 36th International Conference on Machine Learning (ICML) (2019) 3301."},{"key":"24","doi-asserted-by":"crossref","unstructured":"[24] Y. LeCun, <i>et al.<\/i>: \u201cGradient-based learning applied to document recognition,\u201d Proc. IEEE <b>86<\/b> (1998) 2278 (DOI: 10.1109\/5.726791).","DOI":"10.1109\/5.726791"},{"key":"25","unstructured":"[25] K. Simonyan and A. Zisserman: \u201cVery deep convolutional networks for large-scale image recognition,\u201d ICLR (2015)."},{"key":"26","doi-asserted-by":"crossref","unstructured":"[26] Y. LeCun, <i>et al.<\/i>: \u201cGradient-based learning applied to document recognition,\u201d Proc. IEEE <b>86<\/b> (1998) 2278 (DOI: 10.1109\/5.726791).","DOI":"10.1109\/5.726791"},{"key":"27","unstructured":"[27] A. Krizhevsky: \u201cLearning multiple layers of features from tiny images,\u201d Techinical Report (2009)."},{"key":"28","doi-asserted-by":"crossref","unstructured":"[28] J. Deng, <i>et al.<\/i>: \u201cImageNet: a large-scale hierarchical image database,\u201d CVPR (2009) 248 (DOI: 10.1109\/CVPR.2009.5206848).","DOI":"10.1109\/CVPR.2009.5206848"},{"key":"29","unstructured":"[29] S.K. Esser, <i>et al.<\/i>: \u201cLearned step size quantization,\u201d ICLR (2020)."},{"key":"30","unstructured":"[30] A. Paszke, <i>et al.<\/i>: \u201cPyTorch: an imperative style, high-performance deep learning library,\u201d NeurIPS (2019) 8024."},{"key":"31","doi-asserted-by":"crossref","unstructured":"[31] S. Li, <i>et al.<\/i>: \u201cCACTI-P: architecture-level modeling for SRAM-based structures with advanced leakage reduction techniques,\u201d ICCAD (2011) 694 (DOI: 10.1109\/ICCAD.2011.6105405).","DOI":"10.1109\/ICCAD.2011.6105405"}],"container-title":["IEICE Electronics Express"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.jstage.jst.go.jp\/article\/elex\/19\/16\/19_19.20220229\/_pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,8,27]],"date-time":"2022-08-27T04:27:20Z","timestamp":1661574440000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.jstage.jst.go.jp\/article\/elex\/19\/16\/19_19.20220229\/_article"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,8,25]]},"references-count":31,"journal-issue":{"issue":"16","published-print":{"date-parts":[[2022]]}},"URL":"https:\/\/doi.org\/10.1587\/elex.19.20220229","relation":{},"ISSN":["1349-2543"],"issn-type":[{"value":"1349-2543","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,8,25]]},"article-number":"19.20220229"}}