{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,2]],"date-time":"2026-04-02T07:58:42Z","timestamp":1775116722215,"version":"3.50.1"},"reference-count":44,"publisher":"Springer Science and Business Media LLC","issue":"7708","license":[{"start":{"date-parts":[[2018,6,1]],"date-time":"2018-06-01T00:00:00Z","timestamp":1527811200000},"content-version":"tdm","delay-in-days":0,"URL":"http:\/\/www.springer.com\/tdm"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Nature"],"published-print":{"date-parts":[[2018,6]]},"DOI":"10.1038\/s41586-018-0180-5","type":"journal-article","created":{"date-parts":[[2018,5,25]],"date-time":"2018-05-25T14:08:55Z","timestamp":1527257335000},"page":"60-67","update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":950,"title":["Equivalent-accuracy accelerated neural-network training using analogue memory"],"prefix":"10.1038","volume":"558","author":[{"given":"Stefano","family":"Ambrogio","sequence":"first","affiliation":[]},{"given":"Pritish","family":"Narayanan","sequence":"additional","affiliation":[]},{"given":"Hsinyu","family":"Tsai","sequence":"additional","affiliation":[]},{"given":"Robert M.","family":"Shelby","sequence":"additional","affiliation":[]},{"given":"Irem","family":"Boybat","sequence":"additional","affiliation":[]},{"given":"Carmelo","family":"di Nolfo","sequence":"additional","affiliation":[]},{"given":"Severin","family":"Sidler","sequence":"additional","affiliation":[]},{"given":"Massimo","family":"Giordano","sequence":"additional","affiliation":[]},{"given":"Martina","family":"Bodini","sequence":"additional","affiliation":[]},{"given":"Nathan C. P.","family":"Farinha","sequence":"additional","affiliation":[]},{"given":"Benjamin","family":"Killeen","sequence":"additional","affiliation":[]},{"given":"Christina","family":"Cheng","sequence":"additional","affiliation":[]},{"given":"Yassine","family":"Jaoudi","sequence":"additional","affiliation":[]},{"given":"Geoffrey W.","family":"Burr","sequence":"additional","affiliation":[]}],"member":"297","published-online":{"date-parts":[[2018,6,6]]},"reference":[{"key":"180_CR1","doi-asserted-by":"publisher","first-page":"436","DOI":"10.1038\/nature14539","volume":"521","author":"Y LeCun","year":"2015","unstructured":"LeCun, Y., Bengio, Y. & Hinton, G. Deep learning. Nature 521, 436\u2013444 (2015).","journal-title":"Nature"},{"key":"180_CR2","unstructured":"Coates, A. et al. Deep learning with COTS HPC systems. In Proc. 30th International Conference on Machine Learning 1337\u20131345 (Association for Computing Machinery, 2013)."},{"key":"180_CR3","unstructured":"Gupta, S., Agrawal, A., Gopalakrishnan, K. & Narayanan, P. Deep learning with limited numerical precision. In Proc. 30th International Conference on Machine Learning 1737\u20131746 (Association for Computing Machinery, 2015)."},{"key":"180_CR4","unstructured":"Merolla, P., Appuswamy, R., Arthur, J., Esser, S. K. & Modha, D. Deep neural networks are robust to weight binarization and other non-linear distortions. Preprint at \n                    https:\/\/arxiv.org\/abs\/1606.01981\n                    \n                   (2016)."},{"key":"180_CR5","doi-asserted-by":"crossref","unstructured":"Nurvitadhi, E. et al. Can FPGAs beat GPUs in accelerating next-generation deep neural networks? In Proc. 2017 ACM\/SIGSA International Symposium of Field-Programmable Gate Arrays 5\u201314 (Association for Computing Machinery, 2017).","DOI":"10.1145\/3020078.3021740"},{"key":"180_CR6","unstructured":"Jouppi, N. P. et al. In-datacenter performance analysis of a tensor processing unit. In Proc. 2017 International Symposium on Computer Architecture 1\u201312 (Association for Computing Machinery, 2017)."},{"key":"180_CR7","doi-asserted-by":"publisher","first-page":"668","DOI":"10.1126\/science.1254642","volume":"345","author":"PA Merolla","year":"2014","unstructured":"Merolla, P. A. et al. A million spiking-neuron integrated circuit with a scalable communication network and interface. Science 345, 668\u2013673 (2014).","journal-title":"Science"},{"key":"180_CR8","doi-asserted-by":"publisher","first-page":"11441","DOI":"10.1073\/pnas.1604850113","volume":"113","author":"SK Esser","year":"2016","unstructured":"Esser, S. K. et al. Convolutional networks for fast, energy-efficient neuromorphic computing. Proc. Natl Acad. Sci. USA 113, 11441\u201311446 (2016).","journal-title":"Proc. Natl Acad. Sci. USA"},{"key":"180_CR9","doi-asserted-by":"publisher","first-page":"1086","DOI":"10.1109\/4.309904","volume":"29","author":"T Morie","year":"1994","unstructured":"Morie, T. & Amemiya, Y. An all-analog expandable neural network LSI with on-chip backpropagation learning. IEEE J. Solid-State Circuits 29, 1086\u20131093 (1994).","journal-title":"IEEE J. Solid-State Circuits"},{"key":"180_CR10","doi-asserted-by":"crossref","unstructured":"Burr, G. W. et al. Experimental demonstration and tolerancing of a large-scale neural network (165,000 synapses), using phase-change memory as the synaptic weight element. In 2014 IEEE International Electron Devices Meeting T29.5 (IEEE, 2014).","DOI":"10.1109\/IEDM.2014.7047135"},{"key":"180_CR11","doi-asserted-by":"publisher","first-page":"3498","DOI":"10.1109\/TED.2015.2439635","volume":"62","author":"GW Burr","year":"2015","unstructured":"Burr, G. W. et al. Experimental demonstration and tolerancing of a large-scale neural network (165,000 synapses), using phase-change memory as the synaptic weight element. IEEE Trans. Electron Dev. 62, 3498\u20133507 (2015).","journal-title":"IEEE Trans. Electron Dev."},{"key":"180_CR12","doi-asserted-by":"publisher","first-page":"333","DOI":"10.3389\/fnins.2016.00333","volume":"10","author":"T Gokmen","year":"2016","unstructured":"Gokmen, T. & Vlasov, Y. Acceleration of deep neural network training with resistive cross-point devices: design considerations. Front. Neurosci. 10, 333 (2016).","journal-title":"Front. Neurosci."},{"key":"180_CR13","first-page":"89","volume":"2","author":"GW Burr","year":"2017","unstructured":"Burr, G. W. et al. Neuromorphic computing using non-volatile memory. Adv. Physics X 2, 89\u2013124 (2017).","journal-title":"Adv. Physics X"},{"key":"180_CR14","doi-asserted-by":"crossref","unstructured":"Yu, S. et al. Scaling-up resistive synaptic arrays for neuro-inspired architecture: challenges and prospect. In 2015 IEEE International Electron Devices Meeting 17.3 (IEEE, 2015).","DOI":"10.1109\/IEDM.2015.7409718"},{"key":"180_CR15","doi-asserted-by":"publisher","first-page":"455204","DOI":"10.1088\/0957-4484\/26\/45\/455204","volume":"26","author":"L Gao","year":"2015","unstructured":"Gao, L. et al. Fully parallel write\/read in resistive synaptic array for accelerating on-chip learning. Nanotechnology 26, 455204 (2015).","journal-title":"Nanotechnology"},{"key":"180_CR16","doi-asserted-by":"publisher","first-page":"61","DOI":"10.1038\/nature14441","volume":"521","author":"M Prezioso","year":"2015","unstructured":"Prezioso, M. et al. Training and operation of an integrated neuromorphic network based on metal-oxide memristors. Nature 521, 61\u201364 (2015).","journal-title":"Nature"},{"key":"180_CR17","doi-asserted-by":"publisher","first-page":"457","DOI":"10.1109\/LED.2015.2418342","volume":"36","author":"J-W Jang","year":"2015","unstructured":"Jang, J.-W., Park, S., Burr, G. W., Hwang, H. & Jeong, Y.-H. Optimization of conductance change in Pr1\u2212xCa\n                    x\n                  MnO3-based synaptic devices for neuromorphic systems. IEEE Electron Device Lett. 36, 457\u2013459 (2015).","journal-title":"IEEE Electron Device Lett."},{"key":"180_CR18","doi-asserted-by":"publisher","first-page":"173105","DOI":"10.1063\/1.4934818","volume":"107","author":"YJ Jeong","year":"2015","unstructured":"Jeong, Y. J., Kim, S. & Lu, W. D. Utilizing multiple state variables to improve the dynamic range of analog switching in a memristor. Appl. Phys. Lett. 107, 173105 (2015).","journal-title":"Appl. Phys. Lett."},{"key":"180_CR19","doi-asserted-by":"publisher","first-page":"2827","DOI":"10.1109\/TED.2014.2331707","volume":"61","author":"Y Kaneko","year":"2014","unstructured":"Kaneko, Y., Nishitani, Y. & Ueda, M. Ferroelectric artificial synapses for recognition of a multishaded image. IEEE Trans. Electron Dev. 61, 2827\u20132833 (2014).","journal-title":"IEEE Trans. Electron Dev."},{"key":"180_CR20","unstructured":"Nandakumar, S. R. et al. Mixed-precision training of deep neural networks using computational memory. Preprint at \n                    https:\/\/arxiv.org\/abs\/1712.01192\n                    \n                   (2017)."},{"key":"180_CR21","doi-asserted-by":"publisher","first-page":"414","DOI":"10.1038\/nmat4856","volume":"16","author":"Y Burgt van de","year":"2017","unstructured":"van de Burgt, Y. et al. A non-volatile organic electrochemical device as a low-voltage artificial synapse for neuromorphic computing. Nat. Mater. 16, 414\u2013418 (2017).","journal-title":"Nat. Mater."},{"key":"180_CR22","doi-asserted-by":"crossref","unstructured":"Agarwal, S. et al. Achieving ideal accuracies in analog neuromorphic computing using periodic carry. In 2017 Symposium on VLSI Technology T13.2 (IEEE, 2017).","DOI":"10.23919\/VLSIT.2017.7998164"},{"key":"180_CR23","doi-asserted-by":"publisher","first-page":"2278","DOI":"10.1109\/5.726791","volume":"86","author":"Y Lecun","year":"1998","unstructured":"Lecun, Y., Bottou, L., Bengio, Y. & Haffner, P. Gradient-based learning applied to document recognition. Proc. IEEE 86, 2278\u20132324 (1998).","journal-title":"Proc. IEEE"},{"key":"180_CR24","unstructured":"Krizhevsky, A. Learning Multiple Layers of Features From Tiny Images. Ch. 3, \n                    https:\/\/www.cs.toronto.edu\/~kriz\/cifar.html\n                    \n                   (2009)."},{"key":"180_CR25","doi-asserted-by":"publisher","first-page":"11","DOI":"10.1147\/JRD.2017.2716579","volume":"61","author":"P Narayanan","year":"2017","unstructured":"Narayanan, P. et al. Towards on-chip acceleration of the backpropagation algorithm using non-volatile memory. IBM J. Res. Develop. 61, 11 (2017).","journal-title":"IBM J. Res. Develop."},{"key":"180_CR26","doi-asserted-by":"publisher","first-page":"533","DOI":"10.1038\/323533a0","volume":"323","author":"DE Rumelhart","year":"1986","unstructured":"Rumelhart, D. E., Hinton, G. E. & Williams, R. J. Learning representations by backpropagating errors. Nature 323, 533\u2013536 (1986).","journal-title":"Nature"},{"key":"180_CR27","doi-asserted-by":"publisher","first-page":"126","DOI":"10.1016\/j.procs.2014.11.094","volume":"41","author":"Z Xu","year":"2014","unstructured":"Xu, Z. et al. Parallel programming of resistive cross-point array for synaptic plasticity. Procedia Comput. Sci. 41, 126\u2013133 (2014).","journal-title":"Procedia Comput. Sci."},{"key":"180_CR28","doi-asserted-by":"crossref","unstructured":"Papandreou, N. et al. Programming algorithms for multilevel phase-change memory. In 2011 IEEE International Symposium on Circuits and Systems 329\u2013332 (IEEE, 2011).","DOI":"10.1109\/ISCAS.2011.5937569"},{"key":"180_CR29","doi-asserted-by":"publisher","first-page":"075201","DOI":"10.1088\/0957-4484\/23\/7\/075201","volume":"23","author":"F Alibart","year":"2012","unstructured":"Alibart, F., Gao, L., Hoskins, B. D. & Strukov, D. B. High-precision tuning of state for memristive devices by adaptable variation-tolerant algorithm. Nanotechnology 23, 075201 (2012).","journal-title":"Nanotechnology"},{"key":"180_CR30","doi-asserted-by":"crossref","unstructured":"Hu, M. et al. Dot-product engine for neuromorphic computing: programming 1T1M crossbar to accelerate matrix-vector multiplication. In Proc. 53rd Annual Design Automation Conference 19 (Association for Computing Machinery, 2016).","DOI":"10.1145\/2897937.2898010"},{"key":"180_CR31","doi-asserted-by":"publisher","first-page":"1604310","DOI":"10.1002\/adma.201604310","volume":"29","author":"EJ Fuller","year":"2017","unstructured":"Fuller, E. J. et al. Li-ion synaptic transistor for low power analog computing. Adv. Mater. 29, 1604310 (2017).","journal-title":"Adv. Mater."},{"key":"180_CR32","doi-asserted-by":"crossref","unstructured":"Kim, S., Gokmen, T., Lee, H.-M. & Haensch, W. E. Analog CMOS-based resistive processing unit for deep neural network training. In 2017 IEEE 60th International Midwest Symposium on Circuits and Systems 422\u2013425 (IEEE, 2017).","DOI":"10.1109\/MWSCAS.2017.8052950"},{"key":"180_CR33","unstructured":"Goodfellow, I., Bengio, Y. & Courville, A. Deep Learning Ch. 8 (MIT Press, 2016)."},{"key":"180_CR34","unstructured":"Donahue, J. et al. DeCAF: a deep convolutional activation feature for generic visual recognition. In Proc. 31st International Conference on Machine Learning 647\u2013655 (Association for Computing Machinery, 2014)."},{"key":"180_CR35","doi-asserted-by":"crossref","unstructured":"Szegedy, C., Vanhoucke, V., Ioffe, S., Shlens, J. & Wojna, Z. Rethinking the Inception architecture for computer vision. Preprint at \n                    https:\/\/arxiv.org\/abs\/1512.00567\n                    \n                   (2015).","DOI":"10.1109\/CVPR.2016.308"},{"key":"180_CR36","unstructured":"Mujtaba, H. Nvidia Volta GV100 12nm FinFET GPU detailed \u2013 Tesla V100 specifications include 21 billion transistors, 5120 CUDA cores, 16 GB HBM2 with 900 GB\/s bandwidth. Wccftech \n                    https:\/\/wccftech.com\/nvidia-volta-gv100-gpu-tesla-v100-architecture-specifications-deep-dive\/\n                    \n                   (2017)."},{"key":"180_CR37","doi-asserted-by":"publisher","first-page":"1735","DOI":"10.1162\/neco.1997.9.8.1735","volume":"9","author":"S Hochreiter","year":"1997","unstructured":"Hochreiter, S. & Schmidhuber, J. Long short-term memory. Neural Comput. 9, 1735\u20131780 (1997).","journal-title":"Neural Comput."},{"key":"180_CR38","doi-asserted-by":"crossref","unstructured":"Cho, K., van Merrienboer, B., Bahdanau, D. & Bengio, Y. On the properties of neural machine translation: Encoder-decoder approaches. Preprint at \n                    https:\/\/arxiv.org\/abs\/1409.1259\n                    \n                   (2014).","DOI":"10.3115\/v1\/W14-4012"},{"key":"180_CR39","doi-asserted-by":"publisher","first-page":"040802","DOI":"10.1116\/1.4889999","volume":"32","author":"GW Burr","year":"2014","unstructured":"Burr, G. W. et al. Access devices for 3D crosspoint memory. J. Vac. Sci. Technol. B 32, 040802 (2014).","journal-title":"J. Vac. Sci. Technol. B"},{"key":"180_CR40","doi-asserted-by":"crossref","unstructured":"Narayanan, P. et al. Reducing circuit design complexity for neuromorphic machine learning systems based on non-volatile memory arrays. In 2017 IEEE International Symposium on Circuits and Systems 1\u20134 (IEEE, 2017).","DOI":"10.1109\/ISCAS.2017.8050988"},{"key":"180_CR41","doi-asserted-by":"publisher","first-page":"308","DOI":"10.1109\/TED.2006.888752","volume":"54","author":"D Ielmini","year":"2007","unstructured":"Ielmini, D., Lacaita, A. L. & Mantegazza, D. Recovery and drift dynamics of resistance and threshold voltages in phase-change memories. IEEE Trans. Electron Dev. 54, 308\u2013315 (2007).","journal-title":"IEEE Trans. Electron Dev."},{"key":"180_CR42","doi-asserted-by":"publisher","first-page":"1433","DOI":"10.1109\/JSSC.1989.572629","volume":"24","author":"MJM Pelgrom","year":"1989","unstructured":"Pelgrom, M. J. M., Duinmaijer, A. C. J. & Welbers, A. P. G. Matching properties of MOS transistors. IEEE J. Solid-State Circuits 24, 1433\u20131439 (1989).","journal-title":"IEEE J. Solid-State Circuits"},{"key":"180_CR43","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/1862891.1862892","volume":"39","author":"Y Cao","year":"2009","unstructured":"Cao, Y. What is predictive technology model (PTM)? SIGDA Newsl. 39, 1 (2009).","journal-title":"SIGDA Newsl."},{"key":"180_CR44","doi-asserted-by":"crossref","unstructured":"Bengio, Y. Louradour, J. Collobert, R. & Weston, J. Curriculum learning. In Proc. 26th Annual International Conference on Machine Learning 41\u201348 (ACM, 2009).","DOI":"10.1145\/1553374.1553380"}],"container-title":["Nature"],"original-title":[],"language":"en","link":[{"URL":"http:\/\/www.nature.com\/articles\/s41586-018-0180-5","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"http:\/\/www.nature.com\/articles\/s41586-018-0180-5.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"http:\/\/www.nature.com\/articles\/s41586-018-0180-5.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,5,20]],"date-time":"2023-05-20T21:46:35Z","timestamp":1684619195000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.nature.com\/articles\/s41586-018-0180-5"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2018,6]]},"references-count":44,"journal-issue":{"issue":"7708","published-print":{"date-parts":[[2018,6]]}},"alternative-id":["180"],"URL":"https:\/\/doi.org\/10.1038\/s41586-018-0180-5","relation":{},"ISSN":["0028-0836","1476-4687"],"issn-type":[{"value":"0028-0836","type":"print"},{"value":"1476-4687","type":"electronic"}],"subject":[],"published":{"date-parts":[[2018,6]]},"assertion":[{"value":"26 January 2018","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"29 March 2018","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"6 June 2018","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"The authors declare no competing interests.","order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}]}}