{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,13]],"date-time":"2026-03-13T14:56:25Z","timestamp":1773413785994,"version":"3.50.1"},"reference-count":50,"publisher":"Association for Computing Machinery (ACM)","issue":"4","license":[{"start":{"date-parts":[[2021,9,1]],"date-time":"2021-09-01T00:00:00Z","timestamp":1630454400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/www.acm.org\/publications\/policies\/copyright_policy#Background"}],"funder":[{"DOI":"10.13039\/100002570","name":"American Association of Immunologists","doi-asserted-by":"crossref","id":[{"id":"10.13039\/100002570","id-type":"DOI","asserted-by":"crossref"}]},{"name":"College of Engineering, Texas A&M University"}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["J. Emerg. Technol. Comput. Syst."],"published-print":{"date-parts":[[2021,10,31]]},"abstract":"<jats:p>\n            Training deep learning networks involves continuous weight updates across the various layers of the deep network while using a backpropagation (BP) algorithm. This results in expensive computation overheads during training. Consequently, most deep learning accelerators today employ pretrained weights and focus only on improving the design of the inference phase. The recent trend is to build a complete deep learning accelerator by incorporating the training module. Such efforts require an ultra-fast chip architecture for executing the BP algorithm. In this article, we propose a novel photonics-based backpropagation accelerator for high-performance deep learning training. We present the design for a convolutional neural network (CNN),\n            <jats:bold>BPLight-CNN<\/jats:bold>\n            , which incorporates the silicon photonics-based backpropagation accelerator.\n            <jats:bold>BPLight-CNN<\/jats:bold>\n            is a first-of-its-kind photonic and memristor-based CNN architecture for end-to-end training and prediction. We evaluate\n            <jats:bold>BPLight-CNN<\/jats:bold>\n            using a photonic CAD framework (IPKISS) on deep learning benchmark models, including LeNet and VGG-Net. The proposed design achieves (i) at least 34\u00d7 speedup, 34\u00d7 improvement in computational efficiency, and 38.5\u00d7 energy savings during training; and (ii) 29\u00d7 speedup, 31\u00d7 improvement in computational efficiency, and 38.7\u00d7 improvement in energy savings during inference compared with the state-of-the-art designs. All of these comparisons are done at a 16-bit resolution, and BPLight-CNN achieves these improvements at a cost of approximately 6% lower accuracy compared with the state-of-the-art.\n          <\/jats:p>","DOI":"10.1145\/3446212","type":"journal-article","created":{"date-parts":[[2021,9,1]],"date-time":"2021-09-01T18:41:31Z","timestamp":1630521691000},"page":"1-26","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":14,"title":["BPLight-CNN: A Photonics-Based Backpropagation Accelerator for Deep Learning"],"prefix":"10.1145","volume":"17","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-3802-381X","authenticated-orcid":false,"given":"Dharanidhar","family":"Dang","sequence":"first","affiliation":[{"name":"UC San Diego, La Jolla, CA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Sai Vineel Reddy","family":"Chittamuru","sequence":"additional","affiliation":[{"name":"Micron Technology, Austin, Texas"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Sudeep","family":"Pasricha","sequence":"additional","affiliation":[{"name":"Colorado State University, Fort Collins, CO"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Rabi","family":"Mahapatra","sequence":"additional","affiliation":[{"name":"Texas A&amp;M University, College Station, Texas"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Debashis","family":"Sahoo","sequence":"additional","affiliation":[{"name":"UC San Diego, La Jolla, CA"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"320","published-online":{"date-parts":[[2021,9]]},"reference":[{"key":"e_1_2_1_1_1","doi-asserted-by":"publisher","DOI":"10.1145\/2966986.2967011"},{"key":"e_1_2_1_2_1","doi-asserted-by":"publisher","DOI":"10.1145\/3007787.3001139"},{"key":"e_1_2_1_3_1","doi-asserted-by":"crossref","unstructured":"T. Gokmen and Y. Vlasov. 2016. Acceleration of deep neural network training with resistive cross-point devices: Design considerations. Frontiers in Neuroscience 10 (July 2016) 333.  T. Gokmen and Y. Vlasov. 2016. Acceleration of deep neural network training with resistive cross-point devices: Design considerations. Frontiers in Neuroscience 10 (July 2016) 333.","DOI":"10.3389\/fnins.2016.00333"},{"key":"e_1_2_1_4_1","doi-asserted-by":"publisher","DOI":"10.1109\/HPCA.2017.55"},{"key":"e_1_2_1_5_1","doi-asserted-by":"publisher","DOI":"10.3389\/fnins.2017.00538"},{"key":"e_1_2_1_6_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-018-04484-2"},{"key":"e_1_2_1_7_1","doi-asserted-by":"publisher","DOI":"10.1145\/2897937.2898010"},{"key":"e_1_2_1_8_1","doi-asserted-by":"publisher","DOI":"10.1109\/TNN.2011.2161771"},{"issue":"7430","key":"e_1_2_1_9_1","first-page":"2017","article-title":"Neuromorphic photonic networks using silicon photonic weight banks","author":"Tait A. N.","year":"2017","unstructured":"A. N. Tait , T. Ferreira de Lima , E. Zhou , A. X. Wu , M. A. Nahmias , B. J. Shastr , and P. R. Prucnal . 2017 . Neuromorphic photonic networks using silicon photonic weight banks . In Scientific Reports 7, Art. No. 7430 , 2017 . A. N. Tait, T. Ferreira de Lima, E. Zhou, A. X. Wu, M. A. Nahmias, B. J. Shastr, and P. R. Prucnal. 2017. Neuromorphic photonic networks using silicon photonic weight banks. In Scientific Reports 7, Art. No. 7430, 2017.","journal-title":"Scientific Reports 7, Art."},{"key":"e_1_2_1_10_1","doi-asserted-by":"publisher","DOI":"10.1364\/CLEO_SI.2019.SM2J.3"},{"key":"e_1_2_1_11_1","first-page":"864","article-title":"Training of photonic neural networks through in situ backpropagation and gradient measurement","volume":"5","author":"Hughes T. W.","year":"2018","unstructured":"T. W. Hughes , M. Minkov , Y. Shi , and S. Fan . 2018 . Training of photonic neural networks through in situ backpropagation and gradient measurement . OSA Publishing Optica 5 , 7 (2018), 864 \u2013 871 . T. W. Hughes, M. Minkov, Y. Shi, and S. Fan. 2018. Training of photonic neural networks through in situ backpropagation and gradient measurement. OSA Publishing Optica 5, 7 (2018), 864\u2013871.","journal-title":"OSA Publishing Optica"},{"key":"e_1_2_1_12_1","doi-asserted-by":"publisher","DOI":"10.1109\/JSTQE.2020.2982990"},{"key":"e_1_2_1_13_1","first-page":"173","volume-title":"IEEE International System-on-Chip Conference (SOCC) Arlington, USA","author":"Mehrabian A.","year":"2018","unstructured":"A. Mehrabian , Y. Al-Kabani , V. J. Sorger , and T. El-Ghazawi . 2018. PCNNA: A photonic convolutional neural network accelerator . IEEE International System-on-Chip Conference (SOCC) Arlington, USA , pp. 169\u2013 173 , 2018 . A. Mehrabian, Y. Al-Kabani, V. J. Sorger, and T. El-Ghazawi. 2018. PCNNA: A photonic convolutional neural network accelerator. IEEE International System-on-Chip Conference (SOCC) Arlington, USA, pp. 169\u2013173, 2018."},{"key":"e_1_2_1_14_1","doi-asserted-by":"publisher","DOI":"10.1038\/nphoton.2017.93"},{"key":"e_1_2_1_15_1","first-page":"123","volume-title":"IEEE International Conference on High Performance Computing (HiPC) Jaipur","author":"Dang D.","year":"2017","unstructured":"D. Dang , J. Dass , and R. Mahapatra . 2017. ConvLight: A convolutional accelerator with memristor integrated photonic computing . IEEE International Conference on High Performance Computing (HiPC) Jaipur , India , pp. 114\u2013 123 , Feb. 2017 . D. Dang, J. Dass, and R. Mahapatra. 2017. ConvLight: A convolutional accelerator with memristor integrated photonic computing. IEEE International Conference on High Performance Computing (HiPC) Jaipur, India, pp. 114\u2013123, Feb. 2017."},{"key":"e_1_2_1_16_1","doi-asserted-by":"publisher","DOI":"10.5555\/3408352.3408679"},{"key":"e_1_2_1_17_1","first-page":"1488","volume-title":"Automation & Test in Europe Conference & Exhibition (DATE) Florence","author":"Liu W.","year":"2019","unstructured":"W. Liu , W. Liu , Y. Ye , Q. Lou , Y. Xie and L. Jiang . 2019. HolyLight: A nanophotonic accelerator for deep learning in data centers. IEEE\/ACM Design , Automation & Test in Europe Conference & Exhibition (DATE) Florence , Italy , 2019 , pp. 1483\u2013 1488 , DOI:10.23919\/DATE.2019.8715195 10.23919\/DATE.2019.8715195 W. Liu, W. Liu, Y. Ye, Q. Lou, Y. Xie and L. Jiang. 2019. HolyLight: A nanophotonic accelerator for deep learning in data centers. IEEE\/ACM Design, Automation & Test in Europe Conference & Exhibition (DATE) Florence, Italy, 2019, pp. 1483\u20131488, DOI:10.23919\/DATE.2019.8715195"},{"key":"e_1_2_1_18_1","volume-title":"Art. No.","author":"Long Y.","year":"2021","unstructured":"Y. Long , L. Zhou , and Jian Wang. 2016. Photonic-assisted microwave signal multiplication and modulation using a silicon Mach\u2013Zehnder modulator. Scientific Reports, 6 , Art. No. 2021 5, Feb. 2016. Y. Long, L. Zhou, and Jian Wang. 2016. Photonic-assisted microwave signal multiplication and modulation using a silicon Mach\u2013Zehnder modulator. Scientific Reports, 6, Art. No. 20215, Feb. 2016."},{"key":"e_1_2_1_19_1","volume-title":"2002 edition. Boston","year":"2002","unstructured":"Semiconductor Optical Amplifiers , 2002 edition. Boston ; London : Springer , 2002 . Semiconductor Optical Amplifiers, 2002 edition. Boston; London: Springer, 2002."},{"key":"e_1_2_1_20_1","doi-asserted-by":"crossref","unstructured":"P. Li X. Yi X. Liu D. Zhao Y. Zhao and Y. Wang. 2016. All-optical analog comparator. Nature Scientific Reports 6 Art. No. 31903 Aug. 2016.  P. Li X. Yi X. Liu D. Zhao Y. Zhao and Y. Wang. 2016. All-optical analog comparator. Nature Scientific Reports 6 Art. No. 31903 Aug. 2016.","DOI":"10.1038\/srep31903"},{"key":"e_1_2_1_21_1","volume-title":"[Online]. Available: http:\/\/yann.lecun.com\/exdb\/mnist\/","author":"Database NIST","year":"2018","unstructured":"The M NIST Database . ( 2018 ). [Online]. Available: http:\/\/yann.lecun.com\/exdb\/mnist\/ . The MNIST Database. (2018). [Online]. Available: http:\/\/yann.lecun.com\/exdb\/mnist\/."},{"key":"e_1_2_1_22_1","doi-asserted-by":"publisher","DOI":"10.1038\/nature06932"},{"key":"e_1_2_1_23_1","doi-asserted-by":"publisher","DOI":"10.1002\/lpor.201100017"},{"key":"e_1_2_1_24_1","doi-asserted-by":"publisher","DOI":"10.1364\/OE.26.000870"},{"key":"e_1_2_1_25_1","doi-asserted-by":"publisher","DOI":"10.1364\/OE.15.010553"},{"key":"e_1_2_1_26_1","doi-asserted-by":"publisher","DOI":"10.1109\/MDAT.2015.2414417"},{"key":"e_1_2_1_27_1","first-page":"4","volume-title":"IEEE\/ACM International Symposium on Networks-on-Chip (NOCS) Nara","author":"Thakkar I.","year":"2016","unstructured":"I. Thakkar , S. V. R. Chittamuru and S. Pasricha . 2016. Run-Time laser power management in photonic NoCs with on-chip semiconductor optical amplifiers . IEEE\/ACM International Symposium on Networks-on-Chip (NOCS) Nara , Japan , pp. 1\u2013 4 , Aug. 2016 . I. Thakkar, S. V. R. Chittamuru and S. Pasricha. 2016. Run-Time laser power management in photonic NoCs with on-chip semiconductor optical amplifiers. IEEE\/ACM International Symposium on Networks-on-Chip (NOCS) Nara, Japan, pp. 1\u20134, Aug. 2016."},{"key":"e_1_2_1_28_1","doi-asserted-by":"publisher","DOI":"10.1145\/2602155"},{"key":"e_1_2_1_29_1","doi-asserted-by":"publisher","DOI":"10.1002\/lpor.201600276"},{"key":"e_1_2_1_30_1","first-page":"65","volume-title":"IEEE Optical Interconnects Conference Santa Fe, USA","author":"Li C.","year":"2012","unstructured":"C. Li , M. Browning , P. V. Gratz , and S. Palermo . 2012. Energy-efficient optical broadcast for nanophotonic networks-on-chip . IEEE Optical Interconnects Conference Santa Fe, USA , pp. 64\u2013 65 , 2012 . C. Li, M. Browning, P. V. Gratz, and S. Palermo. 2012. Energy-efficient optical broadcast for nanophotonic networks-on-chip. IEEE Optical Interconnects Conference Santa Fe, USA, pp. 64\u201365, 2012."},{"key":"e_1_2_1_31_1","doi-asserted-by":"publisher","DOI":"10.1109\/TPDS.2018.2833876"},{"key":"e_1_2_1_32_1","doi-asserted-by":"crossref","unstructured":"J. J. Yang D. B. Strukov and D. R. Stewart. 2012. Memristive devices for computing. Nature Nanotechnology 8 (Dec. 2012) 13\u201324.  J. J. Yang D. B. Strukov and D. R. Stewart. 2012. Memristive devices for computing. Nature Nanotechnology 8 (Dec. 2012) 13\u201324.","DOI":"10.1038\/nnano.2012.240"},{"key":"e_1_2_1_33_1","first-page":"9","volume-title":"IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Boston, USA","author":"Szegedy C.","year":"2015","unstructured":"C. Szegedy , W. Liu , Y. Jia , P. Sermanet , S. Reed , D. Anguelov , D. Erhan , V. Vanhoucke , and A. Rabinovich . 2015. Going deeper with convolutions . IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Boston, USA , pp. 1\u2013 9 , 2015 . C. Szegedy, W. Liu, Y. Jia, P. Sermanet, S. Reed, D. Anguelov, D. Erhan, V. Vanhoucke, and A. Rabinovich. 2015. Going deeper with convolutions. IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Boston, USA, pp. 1\u20139, 2015."},{"key":"e_1_2_1_34_1","first-page":"385","volume-title":"IEEE International Conference on Computer & Communication Engineering Kuala Lumpur","author":"Uddin M. R.","year":"2016","unstructured":"M. R. Uddin , T. K. Siang , N. Munarah , M. Norfauzi , N. Ahmed and M. A. Salam . 2016. Quality Analysis of a Photonic Micro-ring Resonator , IEEE International Conference on Computer & Communication Engineering Kuala Lumpur , Malaysia , pp. 382\u2013 385 , 2016 . M. R. Uddin, T. K. Siang, N. Munarah, M. Norfauzi, N. Ahmed and M. A. Salam. 2016. Quality Analysis of a Photonic Micro-ring Resonator, IEEE International Conference on Computer & Communication Engineering Kuala Lumpur, Malaysia, pp. 382\u2013385, 2016."},{"key":"e_1_2_1_35_1","doi-asserted-by":"publisher","DOI":"10.1109\/NOCS.2012.31"},{"key":"e_1_2_1_36_1","first-page":"24","volume-title":"IEEE Optical Interconnects Conference","author":"Ma Y.","year":"2014","unstructured":"Y. Ma , Z. Xuan , Y. Liu , R. Ding , Y. Li , A. Eu-Jin Lim , G. Lo , T. Baehr-Jones , and M. Hochberg . 2014. Silicon microring based modulator and filter for high speed transmitters at 1310 nm . IEEE Optical Interconnects Conference , San Diego, CA , 2014 , pp. 23\u2013 24 . Y. Ma, Z. Xuan, Y. Liu, R. Ding, Y. Li, A. Eu-Jin Lim, G. Lo, T. Baehr-Jones, and M. Hochberg. 2014. Silicon microring based modulator and filter for high speed transmitters at 1310 nm. IEEE Optical Interconnects Conference, San Diego, CA, 2014, pp. 23\u201324."},{"key":"e_1_2_1_37_1","doi-asserted-by":"publisher","DOI":"10.1364\/OE.21.011869"},{"key":"e_1_2_1_38_1","doi-asserted-by":"publisher","DOI":"10.1145\/2647868.2654889"},{"key":"e_1_2_1_39_1","first-page":"8","volume-title":"OptoElectronics and Communications Conference (OECC) Kyoto","author":"Fujita T.","year":"2013","unstructured":"T. Fujita , Y. Toba , Y. Miyoshi , and M. Ohashi . 2013. Optical analog multiplier based on phase sensitive amplification . OptoElectronics and Communications Conference (OECC) Kyoto , Japan , pp. 1\u2013 8 , Sept. 2013 . T. Fujita, Y. Toba, Y. Miyoshi, and M. Ohashi. 2013. Optical analog multiplier based on phase sensitive amplification. OptoElectronics and Communications Conference (OECC) Kyoto, Japan, pp. 1\u20138, Sept. 2013."},{"key":"e_1_2_1_40_1","doi-asserted-by":"publisher","DOI":"10.1145\/3065386"},{"key":"e_1_2_1_41_1","doi-asserted-by":"publisher","DOI":"10.1109\/TCSI.2011.2107214"},{"key":"e_1_2_1_42_1","first-page":"1","volume-title":"International Conference on Learning Representations (ICLR) San Diego, USA","author":"Simonyan K.","year":"2015","unstructured":"K. Simonyan and A. Zisserman . 2015. Very deep convolutional networks for large-scale image recognition . International Conference on Learning Representations (ICLR) San Diego, USA , pp. 1\u2013 1 , 2015 . K. Simonyan and A. Zisserman. 2015. Very deep convolutional networks for large-scale image recognition. International Conference on Learning Representations (ICLR) San Diego, USA, pp. 1\u20131, 2015."},{"key":"e_1_2_1_43_1","doi-asserted-by":"publisher","DOI":"10.1109\/5.726791"},{"key":"e_1_2_1_44_1","first-page":"2013","volume-title":"IEEE International Symposium on Circuits and Systems (ISCAS) Montreal","author":"Khorami A.","year":"2016","unstructured":"A. Khorami , M. B. Dastjerdi and A. F. Ahmadi . 2016. A low-power high-speed comparator for analog to digital converters . IEEE International Symposium on Circuits and Systems (ISCAS) Montreal , pp. 2010\u2013 2013 , 2016 . A. Khorami, M. B. Dastjerdi and A. F. Ahmadi. 2016. A low-power high-speed comparator for analog to digital converters. IEEE International Symposium on Circuits and Systems (ISCAS) Montreal, pp. 2010\u20132013, 2016."},{"key":"e_1_2_1_45_1","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-015-0816-y"},{"key":"e_1_2_1_46_1","volume-title":"[Online]. Available: www.lucedaphotonics.com","year":"2018","unstructured":"IPKISS-Photonic Framework. ( 2018 ) [Online]. Available: www.lucedaphotonics.com . IPKISS-Photonic Framework. (2018) [Online]. Available: www.lucedaphotonics.com."},{"key":"e_1_2_1_48_1","doi-asserted-by":"publisher","DOI":"10.5555\/3306619.3306634"},{"key":"e_1_2_1_49_1","doi-asserted-by":"publisher","DOI":"10.1145\/3194554.3194608"},{"key":"e_1_2_1_50_1","doi-asserted-by":"publisher","DOI":"10.1109\/TCAD.2020.3014184"},{"key":"e_1_2_1_51_1","first-page":"311","volume-title":"IEEE\/ACM ASP-DAC","author":"Dang D.","year":"2017","unstructured":"D. Dang , S. V. R. Chittamuru , R. Mahapatra and S. Pasricha . 2017. Islands of heaters: A novel thermal management framework for photonic NoCs . IEEE\/ACM ASP-DAC , Chiba , 2017 , pp. 306\u2013 311 . D. Dang, S. V. R. Chittamuru, R. Mahapatra and S. Pasricha. 2017. Islands of heaters: A novel thermal management framework for photonic NoCs. IEEE\/ACM ASP-DAC, Chiba, 2017, pp. 306\u2013311."}],"container-title":["ACM Journal on Emerging Technologies in Computing Systems"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3446212","content-type":"unspecified","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3446212","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,17]],"date-time":"2025-06-17T20:47:05Z","timestamp":1750193225000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3446212"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,9]]},"references-count":50,"journal-issue":{"issue":"4","published-print":{"date-parts":[[2021,10,31]]}},"alternative-id":["10.1145\/3446212"],"URL":"https:\/\/doi.org\/10.1145\/3446212","relation":{},"ISSN":["1550-4832","1550-4840"],"issn-type":[{"value":"1550-4832","type":"print"},{"value":"1550-4840","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,9]]},"assertion":[{"value":"2020-04-01","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2021-12-01","order":1,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2021-09-01","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}