{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,20]],"date-time":"2026-05-20T08:37:49Z","timestamp":1779266269554,"version":"3.51.4"},"reference-count":45,"publisher":"MIT Press","issue":"3","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Neural Computation"],"published-print":{"date-parts":[[2017,3]]},"abstract":"<jats:p>Recent work in computer science has shown the power of deep learning driven by the backpropagation algorithm in networks of artificial neurons. But real neurons in the brain are different from most of these artificial ones in at least three crucial ways: they emit spikes rather than graded outputs, their inputs and outputs are related dynamically rather than by piecewise-smooth functions, and they have no known way to coordinate arrays of synapses in separate forward and feedback pathways so that they change simultaneously and identically, as they do in backpropagation. Given these differences, it is unlikely that current deep learning algorithms can operate in the brain, but we that show these problems can be solved by two simple devices: learning rules can approximate dynamic input-output relations with piecewise-smooth functions, and a variation on the feedback alignment algorithm can train deep networks without having to coordinate forward and feedback synapses. Our results also show that deep spiking networks learn much better if each neuron computes an intracellular teaching signal that reflects that cell\u2019s nonlinearity. With this mechanism, networks of spiking neurons show useful learning in synapses at least nine layers upstream from the output cells and perform well compared to other spiking networks in the literature on the MNIST digit recognition task.<\/jats:p>","DOI":"10.1162\/neco_a_00929","type":"journal-article","created":{"date-parts":[[2017,1,17]],"date-time":"2017-01-17T21:43:31Z","timestamp":1484689411000},"page":"578-602","source":"Crossref","is-referenced-by-count":60,"title":["Deep Learning with Dynamic Spiking Neurons and Fixed Feedback Weights"],"prefix":"10.1162","volume":"29","author":[{"given":"Arash","family":"Samadi","sequence":"first","affiliation":[{"name":"Department of Physiology, University of Toronto, Toronto, Ontario, M5S 1A8, Canada"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Timothy P.","family":"Lillicrap","sequence":"additional","affiliation":[{"name":"Google DeepMind, London, EC4A 3TW, U.K."}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Douglas B.","family":"Tweed","sequence":"additional","affiliation":[{"name":"Department of Physiology, University of Toronto, Toronto, Ontario, M5S 1A8, Canada, and Centre for Vision Research, York University, Toronto, Ontario, M3J 1PC, Canada"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"281","reference":[{"key":"B1","doi-asserted-by":"publisher","DOI":"10.1109\/JPROC.2014.2314454"},{"key":"B2","first-page":"2654","volume":"27","author":"Ba L. J.","year":"2014","journal-title":"Advances in neural information processing systems"},{"key":"B3","doi-asserted-by":"publisher","DOI":"10.1561\/2200000006"},{"key":"B4","doi-asserted-by":"crossref","first-page":"321","DOI":"10.7551\/mitpress\/7496.003.0016","volume-title":"Large-scale kernel machines","author":"Bengio Y.","year":"2007"},{"key":"B5","doi-asserted-by":"publisher","DOI":"10.1016\/j.neunet.2013.07.012"},{"key":"B6","doi-asserted-by":"publisher","DOI":"10.1016\/S0925-2312(01)00658-0"},{"key":"B7","doi-asserted-by":"publisher","DOI":"10.1162\/neco.2007.19.11.2881"},{"key":"B8","doi-asserted-by":"publisher","DOI":"10.1162\/NECO_a_00277"},{"key":"B9","doi-asserted-by":"publisher","DOI":"10.1162\/NECO_a_00052"},{"key":"B10","doi-asserted-by":"publisher","DOI":"10.1038\/337129a0"},{"key":"B11","doi-asserted-by":"publisher","DOI":"10.1016\/j.tins.2008.09.007"},{"key":"B12","doi-asserted-by":"publisher","DOI":"10.3389\/fncom.2015.00099"},{"key":"B13","author":"Diehl P. U.","year":"2015","journal-title":"Fast-classifying, high-accuracy spiking deep networks through weight and threshold balancing"},{"key":"B14","author":"Eliasmith C.","year":"2002","journal-title":"Neural engineering"},{"key":"B15","doi-asserted-by":"publisher","DOI":"10.1126\/science.1225266"},{"key":"B16","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2012.231"},{"issue":"10","key":"B17","first-page":"658","volume":"62","author":"Fukushima K.","year":"1979","journal-title":"Electron. & Commun. Japan"},{"key":"B18","doi-asserted-by":"publisher","DOI":"10.1016\/j.neunet.2012.09.016"},{"key":"B19","doi-asserted-by":"publisher","DOI":"10.1016\/S0364-0213(87)80025-3"},{"key":"B20","author":"Henderson J. A.","year":"2015","journal-title":"Spike event based learning in neural networks"},{"key":"B22","doi-asserted-by":"publisher","DOI":"10.1162\/neco.2006.18.7.1527"},{"key":"B23","doi-asserted-by":"publisher","DOI":"10.1126\/science.1127647"},{"key":"B24","doi-asserted-by":"publisher","DOI":"10.3389\/fncom.2014.00038"},{"key":"B25","author":"Koch C.","year":"1999","journal-title":"Biophysics of computation"},{"key":"B26","author":"Kolen J. F.","year":"1994","journal-title":"Neural Networks 1994: IEEE World Conference on Computational Intelligence"},{"key":"B27","first-page":"1097","volume-title":"Advances in neural information processing systems","volume":"25","author":"Krizhevsky A.","year":"2012"},{"key":"B28","doi-asserted-by":"publisher","DOI":"10.1109\/5.726791"},{"key":"B29","doi-asserted-by":"publisher","DOI":"10.4324\/9781410605504"},{"key":"B30","author":"Lillicrap T. P.","year":"2014","journal-title":"Random feedback weights support learning in deep neural networks"},{"key":"B31","doi-asserted-by":"publisher","DOI":"10.1152\/jn.1994.72.2.974"},{"key":"B32","doi-asserted-by":"publisher","DOI":"10.1002\/9780470608593"},{"key":"B33","doi-asserted-by":"publisher","DOI":"10.1016\/j.jcss.2004.04.001"},{"key":"B34","doi-asserted-by":"publisher","DOI":"10.1038\/81486"},{"key":"B35","doi-asserted-by":"publisher","DOI":"10.3389\/fnins.2013.00272"},{"key":"B36","doi-asserted-by":"publisher","DOI":"10.1109\/TVLSI.2013.2294916"},{"key":"B37","doi-asserted-by":"publisher","DOI":"10.3389\/fnins.2013.00178"},{"key":"B38","author":"Rolls E.","year":"2002","journal-title":"The computational neuroscience of vision"},{"key":"B39","doi-asserted-by":"publisher","DOI":"10.1113\/jphysiol.1987.sp016881"},{"key":"B40","author":"Saxe A.","year":"2013","journal-title":"Proceedings of the 35th Annual Meeting of the Cognitive Science Society"},{"key":"B41","doi-asserted-by":"publisher","DOI":"10.1016\/j.neunet.2014.09.003"},{"key":"B42","author":"Sermanet P.","year":"2013","journal-title":"Over-Feat: Integrated recognition, localization and detection using convolutional networks"},{"key":"B43","first-page":"1929","volume":"15","author":"Srivastava N.","year":"2014","journal-title":"Journal of Machine Learning Research"},{"key":"B44","first-page":"241","author":"Stork D.","year":"1989","journal-title":"1989 IEEE INNS International Joint Conference on Neural Networks"},{"key":"B45","author":"Sutskever I.","year":"2013","journal-title":"Training recurrent neural networks."},{"key":"B46","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.1403112111"}],"container-title":["Neural Computation"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mitpressjournals.org\/doi\/pdf\/10.1162\/NECO_a_00929","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,8,21]],"date-time":"2023-08-21T20:06:51Z","timestamp":1692648411000},"score":1,"resource":{"primary":{"URL":"https:\/\/direct.mit.edu\/neco\/article\/29\/3\/578-602\/8251"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2017,3]]},"references-count":45,"journal-issue":{"issue":"3","published-print":{"date-parts":[[2017,3]]}},"alternative-id":["10.1162\/NECO_a_00929"],"URL":"https:\/\/doi.org\/10.1162\/neco_a_00929","relation":{},"ISSN":["0899-7667","1530-888X"],"issn-type":[{"value":"0899-7667","type":"print"},{"value":"1530-888X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2017,3]]}}}