{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,3]],"date-time":"2026-07-03T16:59:29Z","timestamp":1783097969929,"version":"3.54.6"},"reference-count":42,"publisher":"Association for Computing Machinery (ACM)","issue":"3","license":[{"start":{"date-parts":[[2022,1,29]],"date-time":"2022-01-29T00:00:00Z","timestamp":1643414400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/www.acm.org\/publications\/policies\/copyright_policy#Background"}],"funder":[{"DOI":"10.13039\/501100001711","name":"SNSF","doi-asserted-by":"crossref","award":["CRSII5 177255"],"award-info":[{"award-number":["CRSII5 177255"]}],"id":[{"id":"10.13039\/501100001711","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/100014553","name":"Samsung Advanced Institute of Technology","doi-asserted-by":"crossref","id":[{"id":"10.13039\/100014553","id-type":"DOI","asserted-by":"crossref"}]},{"name":"National Science Foundation Graduate Research Fellowship","award":["DGE 1752814"],"award-info":[{"award-number":["DGE 1752814"]}]}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["J. Emerg. Technol. Comput. Syst."],"published-print":{"date-parts":[[2022,7,31]]},"abstract":"<jats:p>Spiking Neural Networks (SNNs) is a promising paradigm for efficient event-driven processing of spatio-temporally sparse data streams. Spiking Neural Networks (SNNs) have inspired the design of and can take advantage of the emerging class of neuromorphic processors like Intel Loihi. These novel hardware architectures expose a variety of constraints that affect firmware, compiler, and algorithm development alike. To enable rapid and flexible development of SNN algorithms on Loihi, we developed NxTF: a programming interface derived from Keras and compiler optimized for mapping deep convolutional SNNs to the multi-core Intel Loihi architecture. We evaluate NxTF on Deep Neural Networks (DNNs) trained directly on spikes as well as models converted from traditional DNNs, processing both sparse event-based and dense frame-based datasets. Further, we assess the effectiveness of the compiler to distribute models across a large number of cores and to compress models by exploiting Loihi\u2019s weight-sharing features. Finally, we evaluate model accuracy, energy, and time-to-solution compared to other architectures. The compiler achieves near-optimal resource utilization of 80% across 16 Loihi chips for a 28-layer, 4M parameter MobileNet model with input size 128\u00d7128. In addition, we report the lowest error rate of 8.52% for the CIFAR-10 dataset on neuromorphic hardware, using an off-the-shelf MobileNet.<\/jats:p>","DOI":"10.1145\/3501770","type":"journal-article","created":{"date-parts":[[2022,1,29]],"date-time":"2022-01-29T13:39:17Z","timestamp":1643463557000},"page":"1-22","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":51,"title":["NxTF: An API and Compiler for Deep Spiking Neural Networks on Intel Loihi"],"prefix":"10.1145","volume":"18","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-1628-707X","authenticated-orcid":false,"given":"Bodo","family":"Rueckauer","sequence":"first","affiliation":[{"name":"Institute of Neuroinformatics, University of Zurich and ETH Zurich, Winterthurerstrasse, Zurich, Switzerland"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Connor","family":"Bybee","sequence":"additional","affiliation":[{"name":"Redwood Center for Theoretical Neuroscience, University of California, Berkeley, California"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ralf","family":"Goettsche","sequence":"additional","affiliation":[{"name":"Intel Corporation, Neubiberg, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yashwardhan","family":"Singh","sequence":"additional","affiliation":[{"name":"Intel Corporation, Hillsboro, Oregon, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Joyesh","family":"Mishra","sequence":"additional","affiliation":[{"name":"Intel Corporation, Hillsboro, Oregon, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Andreas","family":"Wild","sequence":"additional","affiliation":[{"name":"Intel Corporation, Hillsboro, Oregon, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"320","published-online":{"date-parts":[[2022,1,29]]},"reference":[{"key":"e_1_3_2_2_2","doi-asserted-by":"publisher","DOI":"10.1109\/IJCNN.2013.6707078"},{"key":"e_1_3_2_3_2","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.781"},{"key":"e_1_3_2_4_2","first-page":"1","article-title":"Nengo: A python tool for building large-scale functional brain models","volume":"7","author":"Bekolay Trevor","year":"2014","unstructured":"Trevor Bekolay, James Bergstra, Eric Hunsberger, Travis DeWolf, Terrence C. Stewart, Daniel Rasmussen, Xuan Choo, Aaron Russell Voelker, and Chris Eliasmith. 2014. Nengo: A python tool for building large-scale functional brain models. Front. Neuroinform. 7, JAN (2014), 1\u201313.","journal-title":"Front. Neuroinform."},{"key":"e_1_3_2_5_2","first-page":"1","volume-title":"3rd MLSys Conference","author":"Blalock Davis","year":"2020","unstructured":"Davis Blalock, Jose Javier Gonzalez Ortiz, Jonathan Frankle, and John Guttag. 2020. What is the state of neural network pruning? In 3rd MLSys Conference. MLSys, Austin, TX, 1\u201317."},{"key":"e_1_3_2_6_2","doi-asserted-by":"publisher","DOI":"10.1145\/3320288.3320304"},{"issue":"0","key":"e_1_3_2_7_2","first-page":"1","article-title":"cuDNN: Efficient primitives for deep learning","volume":"0","author":"Chetlur Sharan","year":"2014","unstructured":"Sharan Chetlur, Cliff Woolley, Philippe Vandermersch, Jonathan Cohen, John Tran, Bryan Catanzaro, and Evan Shelhamer. 2014. cuDNN: Efficient primitives for deep learning. arXiv [cs.NE] 0, 0 (Oct. 2014), 1\u20139.","journal-title":"arXiv [cs.NE]"},{"key":"e_1_3_2_8_2","unstructured":"Fran\u00e7ois Chollet et\u00a0al. 2015. Keras. GitHub. Retrieved from https:\/\/github.com\/fchollet\/keras."},{"issue":"11","key":"e_1_3_2_9_2","first-page":"1","article-title":"PyNN: A common interface for neuronal network simulators","volume":"2","author":"Davidson Andrew P.","year":"2009","unstructured":"Andrew P. Davidson, Daniel Br\u00fcderle, Jochen M. Eppler, Jens Kremkow, Eilif Muller, Dejan Pecevski, Laurent Perrinet, and Pierre Yger. 2009. PyNN: A common interface for neuronal network simulators. Front. Neuroinform. 2, 11 (2009), 1\u201320.","journal-title":"Front. Neuroinform."},{"key":"e_1_3_2_10_2","doi-asserted-by":"publisher","DOI":"10.1038\/s42256-019-0097-1"},{"key":"e_1_3_2_11_2","doi-asserted-by":"publisher","DOI":"10.1109\/MM.2018.112130359"},{"key":"e_1_3_2_12_2","doi-asserted-by":"publisher","DOI":"10.1109\/JPROC.2021.3067593"},{"key":"e_1_3_2_13_2","doi-asserted-by":"publisher","DOI":"10.3389\/fnbot.2020.568359"},{"key":"e_1_3_2_14_2","first-page":"1","article-title":"Unsupervised learning of digit recognition using spike-timing-dependent plasticity","volume":"9","author":"Diehl Peter U.","year":"2015","unstructured":"Peter U. Diehl and Matthew Cook. 2015. Unsupervised learning of digit recognition using spike-timing-dependent plasticity. Front. Comput. Neurosci. 9, August (Aug. 2015), 1.","journal-title":"Front. Comput. Neurosci."},{"key":"e_1_3_2_15_2","doi-asserted-by":"publisher","DOI":"10.5555\/2969239.2969364"},{"key":"e_1_3_2_16_2","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.1604850113"},{"key":"e_1_3_2_17_2","doi-asserted-by":"publisher","DOI":"10.1145\/3381755.3398695"},{"key":"e_1_3_2_18_2","doi-asserted-by":"publisher","DOI":"10.1109\/TC.2012.142"},{"key":"e_1_3_2_19_2","first-page":"1","article-title":"MobileNets: Efficient convolutional neural networks for mobile vision applications","volume":"1704","author":"Howard Andrew G.","year":"2017","unstructured":"Andrew G. Howard, Menglong Zhu, Bo Chen, Dmitry Kalenichenko, Weijun Wang, Tobias Weyand, Marco Andreetto, and Hartwig Adam. 2017. MobileNets: Efficient convolutional neural networks for mobile vision applications. CoRR abs\/1704.04861 (2017), 1.","journal-title":"CoRR"},{"key":"e_1_3_2_20_2","first-page":"2495","volume-title":"IEEE International Symposium on Circuits and Systems","author":"Kiselev Ilya","year":"2016","unstructured":"Ilya Kiselev, Daniel Neil, and Shih Chii Liu. 2016. Event-driven deep neural network hardware system for sensor fusion. In IEEE International Symposium on Circuits and Systems. IEEE, Montreal, Canada, 2495\u20132498."},{"key":"e_1_3_2_21_2","doi-asserted-by":"publisher","DOI":"10.1162\/neco.1989.1.4.541"},{"key":"e_1_3_2_22_2","doi-asserted-by":"publisher","DOI":"10.1109\/MC.2018.157113521"},{"key":"e_1_3_2_23_2","doi-asserted-by":"publisher","DOI":"10.1109\/IJCNN48605.2020.9207109"},{"key":"e_1_3_2_24_2","doi-asserted-by":"publisher","DOI":"10.1126\/science.1254642"},{"issue":"7","key":"e_1_3_2_25_2","first-page":"3227","article-title":"Supervised learning based on temporal coding in spiking neural networks","volume":"29","author":"Mostafa Hesham","year":"2018","unstructured":"Hesham Mostafa. 2018. Supervised learning based on temporal coding in spiking neural networks. IEEE Transactions on Neural Networks and Learning Systems 29, 7 (2018), 3227\u20133235.","journal-title":"IEEE Transactions on Neural Networks and Learning Systems"},{"key":"e_1_3_2_26_2","first-page":"1","volume-title":"IEEE International Symposium on Circuits and Systems","author":"Mostafa Hesham","year":"2017","unstructured":"Hesham Mostafa, Bruno U. Pedroni, Sadique Sheik, and Gert Cauwenberghs. 2017. Fast classification using sparsely active spiking networks. In IEEE International Symposium on Circuits and Systems. IEEE, Baltimore, MD, 1."},{"key":"e_1_3_2_27_2","doi-asserted-by":"publisher","DOI":"10.3389\/fnins.2015.00437"},{"key":"e_1_3_2_28_2","doi-asserted-by":"publisher","DOI":"10.5555\/3454287.3455008"},{"key":"e_1_3_2_29_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.neunet.2019.09.008"},{"key":"e_1_3_2_30_2","first-page":"1","volume-title":"International Conference on Learning Representations","author":"Rathi Nitin","year":"2020","unstructured":"Nitin Rathi, Gopalakrishnan Srinivasan, Priyadarshini Panda, and Kaushik Roy. 2020. Enabling deep spiking neural networks with hybrid conversion and spike timing dependent backpropagation. In International Conference on Learning Representations. OpenReview.net, Addis Ababa, Ethiopia, 1."},{"key":"e_1_3_2_31_2","doi-asserted-by":"publisher","DOI":"10.1109\/HPEC.2019.8916327"},{"key":"e_1_3_2_32_2","first-page":"1","article-title":"Spynnaker: A software package for running Pynn simulations on spinnaker","volume":"12","author":"Rhodes Oliver","year":"2018","unstructured":"Oliver Rhodes, Petru\u0163 A. Bogdan, Christian Brenninkmeijer, Simon Davidson, Donal Fellows, Andrew Gait, David R. Lester, Mantas Mikaitis, Luis A. Plana, Andrew G. D. Rowley, Alan B. Stokes, and Steve B. Furber. 2018. Spynnaker: A software package for running Pynn simulations on spinnaker. Front. Neurosci. 12, November (2018), 1.","journal-title":"Front. Neurosci."},{"key":"e_1_3_2_33_2","doi-asserted-by":"publisher","DOI":"10.3389\/fnins.2019.00231"},{"key":"e_1_3_2_34_2","first-page":"8","volume-title":"IEEE International Symposium on Circuits and Systems","author":"Rueckauer Bodo","year":"2018","unstructured":"Bodo Rueckauer and Shih Chii Liu. 2018. Conversion of analog to spiking neural networks using sparse temporal coding. In IEEE International Symposium on Circuits and Systems. IEEE, Florence, Italy, 8\u201312."},{"key":"e_1_3_2_35_2","doi-asserted-by":"publisher","DOI":"10.1109\/IJCNN52387.2021.9533837"},{"key":"e_1_3_2_36_2","first-page":"1","article-title":"Conversion of continuous-valued deep networks to efficient event-driven networks for image classification","volume":"11","author":"Rueckauer Bodo","year":"2017","unstructured":"Bodo Rueckauer, Iulia-Alexandra Lungu, Yuhuang Hu, Michael Pfeiffer, and Shih-Chii Liu. 2017. Conversion of continuous-valued deep networks to efficient event-driven networks for image classification. Front. Neurosci. 11, December (2017), 1\u201312.","journal-title":"Front. Neurosci."},{"key":"e_1_3_2_37_2","first-page":"2227","volume-title":"International Joint Conference on Neural Networks","author":"Schmitt Sebastian","year":"2017","unstructured":"Sebastian Schmitt, Johann Klahn, Guillaume Bellec, Andreas Grubl, Maurice Guttler, Andreas Hartel, Stephan Hartmann, Dan Husmann, Kai Husmann, Sebastian Jeltsch, Vitali Karasenko, Mitja Kleider, Christoph Koke, Alexander Kononov, Christian Mauch, Eric Muller, Paul Muller, Johannes Partzsch, Mihai A. Petrovici, Stefan Schiefer, Stefan Scholze, Vasilis Thanasoulis, Bernhard Vogginger, Robert Legenstein, Wolfgang Maass, Christian Mayr, Rene Schuffny, Johannes Schemmel, and Karlheinz Meier. 2017. Neuromorphic hardware in the loop: Training a deep spiking network on the brainscales wafer-scale system. In International Joint Conference on Neural Networks. IEEE, Anchorage, Alaska, 2227\u20132234."},{"key":"e_1_3_2_38_2","doi-asserted-by":"publisher","DOI":"10.5555\/3326943.3327073"},{"key":"e_1_3_2_39_2","doi-asserted-by":"publisher","DOI":"10.3389\/fnins.2020.00662"},{"key":"e_1_3_2_40_2","first-page":"1","volume-title":"International Joint Conference on Neural Networks","author":"Stromatias Evangelos","year":"2015","unstructured":"Evangelos Stromatias, Daniel Neil, Francesco Galluppi, Michael Pfeiffer, Shih Chii Liu, and Steve Furber. 2015. Scalable energy-efficient, low-latency implementations of trained spiking deep belief networks on SpiNNaker. In International Joint Conference on Neural Networks. IEEE, Killarney, Ireland, 1\u20138."},{"key":"e_1_3_2_41_2","doi-asserted-by":"publisher","DOI":"10.1109\/JETCAS.2019.2951121"},{"key":"e_1_3_2_42_2","unstructured":"Qiang Yu Chenxiang Ma Shiming Song Gaoyan Zhang Jianwu Dang and Kay Chen Tan. 2020. Constructing accurate and efficient deep spiking neural networks with double-threshold and augmented schemes. IEEE Transactions on Neural Networks and Learning Systems ."},{"key":"e_1_3_2_43_2","first-page":"1","article-title":"Efficient computation in adaptive artificial spiking neural networks","volume":"12","author":"Zambrano Davide","year":"2019","unstructured":"Davide Zambrano, Roeland Nusselder, H. Steven Scholte, and Sander Bohte. 2019. Efficient computation in adaptive artificial spiking neural networks. Front. Neurosci. 12(January2019), 1\u201311.","journal-title":"Front. Neurosci."}],"container-title":["ACM Journal on Emerging Technologies in Computing Systems"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3501770","content-type":"unspecified","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3501770","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,17]],"date-time":"2025-06-17T20:11:45Z","timestamp":1750191105000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3501770"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,1,29]]},"references-count":42,"journal-issue":{"issue":"3","published-print":{"date-parts":[[2022,7,31]]}},"alternative-id":["10.1145\/3501770"],"URL":"https:\/\/doi.org\/10.1145\/3501770","relation":{},"ISSN":["1550-4832","1550-4840"],"issn-type":[{"value":"1550-4832","type":"print"},{"value":"1550-4840","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,1,29]]},"assertion":[{"value":"2021-01-01","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2021-11-01","order":1,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2022-01-29","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}