{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,27]],"date-time":"2026-03-27T08:23:58Z","timestamp":1774599838049,"version":"3.50.1"},"reference-count":57,"publisher":"IOP Publishing","issue":"2","license":[{"start":{"date-parts":[[2024,6,20]],"date-time":"2024-06-20T00:00:00Z","timestamp":1718841600000},"content-version":"vor","delay-in-days":19,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2024,6,20]],"date-time":"2024-06-20T00:00:00Z","timestamp":1718841600000},"content-version":"tdm","delay-in-days":19,"URL":"https:\/\/iopscience.iop.org\/info\/page\/text-and-data-mining"}],"funder":[{"name":"National Science Foundation","award":["1954589"],"award-info":[{"award-number":["1954589"]}]},{"DOI":"10.13039\/100009238","name":"Virginia Commonwealth University","doi-asserted-by":"crossref","award":["VCU Breakthrough grant: Energy Efficient Neurocomputer"],"award-info":[{"award-number":["VCU Breakthrough grant: Energy Efficient Neurocomputer"]}],"id":[{"id":"10.13039\/100009238","id-type":"DOI","asserted-by":"crossref"}]},{"name":"Commonwealth of Cyber Initiative (CCI) grant","award":["VV-1Q23-008"],"award-info":[{"award-number":["VV-1Q23-008"]}]}],"content-domain":{"domain":["iopscience.iop.org"],"crossmark-restriction":false},"short-container-title":["Neuromorph. Comput. Eng."],"published-print":{"date-parts":[[2024,6,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:p>Anomaly detection in real-time using autoencoders implemented on edge devices is exceedingly challenging due to limited hardware, energy, and computational resources. We show that these limitations can be addressed by designing an autoencoder with low-resolution non-volatile memory-based synapses and employing an effective quantized neural network learning algorithm. We further propose nanoscale ferromagnetic racetracks with engineered notches hosting magnetic domain walls (DW) as exemplary non-volatile memory-based autoencoder synapses, where limited state (5-state) synaptic weights are manipulated by spin orbit torque (SOT) current pulses to write different magnetoresistance states. The performance of anomaly detection of the proposed autoencoder model is evaluated on the NSL-KDD dataset. Limited resolution and DW device stochasticity aware training of the autoencoder is performed, which yields comparable anomaly detection performance to the autoencoder having floating-point precision weights. While the limited number of quantized states and the inherent stochastic nature of DW synaptic weights in nanoscale devices are typically known to negatively impact the performance, our hardware-aware training algorithm is shown to leverage these imperfect device characteristics to generate an improvement in anomaly detection accuracy (90.98%) compared to accuracy obtained with floating-point synaptic weights that are extremely memory intensive. Furthermore, our DW-based approach demonstrates a remarkable reduction of at least three orders of magnitude in weight updates during training compared to the floating-point approach, implying significant reduction in operation energy for our method. This work could stimulate the development of extremely energy efficient non-volatile multi-state synapse-based processors that can perform real-time training and inference on the edge with unsupervised data.<\/jats:p>","DOI":"10.1088\/2634-4386\/ad49ce","type":"journal-article","created":{"date-parts":[[2024,5,11]],"date-time":"2024-05-11T02:02:26Z","timestamp":1715392946000},"page":"024012","update-policy":"https:\/\/doi.org\/10.1088\/crossmark-policy","source":"Crossref","is-referenced-by-count":5,"title":["Quantized non-volatile nanomagnetic domain wall synapse based autoencoder for efficient unsupervised network anomaly detection"],"prefix":"10.1088","volume":"4","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-7520-5308","authenticated-orcid":true,"given":"Muhammad Sabbir","family":"Alam","sequence":"first","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4517-3330","authenticated-orcid":false,"given":"Walid Al","family":"Misba","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5681-0884","authenticated-orcid":true,"given":"Jayasimha","family":"Atulasimha","sequence":"additional","affiliation":[]}],"member":"266","published-online":{"date-parts":[[2024,6,20]]},"reference":[{"key":"ncead49cebib1","doi-asserted-by":"publisher","first-page":"1174","DOI":"10.3390\/s21041174","article-title":"Next-generation internet of things (iot): opportunities, challenges, and solutions","volume":"21","author":"Zikria","year":"2021","journal-title":"Sensors"},{"key":"ncead49cebib2","doi-asserted-by":"publisher","first-page":"973","DOI":"10.1016\/j.jcss.2014.02.005","article-title":"A survey of emerging threats in cybersecurity","volume":"80","author":"Jang-Jaccard","year":"2014","journal-title":"J. Comput. Syst. Sci."},{"key":"ncead49cebib3","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/1541880.1541882","article-title":"Anomaly detection: a survey","volume":"41","author":"Chandola","year":"2009","journal-title":"ACM Comput. Surv."},{"key":"ncead49cebib4","doi-asserted-by":"publisher","first-page":"134","DOI":"10.1016\/j.neucom.2017.04.070","article-title":"Unsupervised real-time anomaly detection for streaming data","volume":"262","author":"Ahmad","year":"2017","journal-title":"Neurocomputing"},{"key":"ncead49cebib5","article-title":"Deep learning for anomaly detection: a survey","author":"Chalapathy","year":"2019"},{"key":"ncead49cebib6","first-page":"665","article-title":"Anomaly detection with robust deep autoencoders","author":"Zhou","year":"2017"},{"key":"ncead49cebib7","doi-asserted-by":"publisher","first-page":"140136","DOI":"10.1109\/ACCESS.2021.3116612","article-title":"Improving performance of autoencoder-based network anomaly detection on nsl-kdd dataset","volume":"9","author":"Xu","year":"2021","journal-title":"IEEE Access"},{"key":"ncead49cebib8","first-page":"20","article-title":"Challenges and opportunities in edge computing","author":"Varghese","year":"2016"},{"key":"ncead49cebib9","doi-asserted-by":"publisher","first-page":"637","DOI":"10.1109\/JIOT.2016.2579198","article-title":"Edge computing: vision and challenges","volume":"3","author":"Shi","year":"2016","journal-title":"IEEE Internet Things J."},{"key":"ncead49cebib10","article-title":"A survey of quantization methods for efficient neural network inference","author":"Gholami","year":"2021"},{"key":"ncead49cebib11","first-page":"6869","article-title":"Quantized neural networks: training neural networks with low precision weights and activations","volume":"18","author":"Hubara","year":"2017","journal-title":"J. Mach. Learn. Res."},{"key":"ncead49cebib12","first-page":"580","article-title":"Value-aware quantization for training and inference of neural networks","author":"Park","year":"2018"},{"key":"ncead49cebib13","article-title":"Training with quantization noise for extreme model compression","author":"Fan","year":"2020"},{"key":"ncead49cebib14","doi-asserted-by":"publisher","first-page":"529","DOI":"10.1038\/s41565-020-0655-z","article-title":"Memory devices and applications for in-memory computing","volume":"15","author":"Sebastian","year":"2020","journal-title":"Nat. Nanotechnol."},{"key":"ncead49cebib15","doi-asserted-by":"publisher","first-page":"333","DOI":"10.1038\/s41928-018-0092-2","article-title":"In-memory computing with resistive switching devices","volume":"1","author":"Ielmini","year":"2018","journal-title":"Nat. Electron."},{"key":"ncead49cebib16","doi-asserted-by":"publisher","first-page":"191","DOI":"10.1038\/nnano.2015.29","article-title":"Memory leads the way to better computing","volume":"10","author":"Wong","year":"2015","journal-title":"Nat. Nanotechnol."},{"key":"ncead49cebib17","doi-asserted-by":"publisher","first-page":"1152","DOI":"10.1109\/TBCAS.2016.2525823","article-title":"Proposal for an all-spin artificial neural network: emulating neural and synaptic functionalities through domain wall motion in ferromagnets","volume":"10","author":"Sengupta","year":"2016","journal-title":"IEEE Trans. Biomed. Circuits Syst."},{"key":"ncead49cebib18","doi-asserted-by":"publisher","DOI":"10.1016\/j.jmmm.2019.165434","article-title":"On-chip learning for domain wall synapse based fully connected neural network","volume":"489","author":"Bhowmik","year":"2019","journal-title":"J. Magn. Magn. Mater."},{"key":"ncead49cebib19","doi-asserted-by":"publisher","first-page":"166","DOI":"10.1109\/TBCAS.2015.2414423","article-title":"Spin-transfer torque magnetic memory as a stochastic memristive synapse for neuromorphic systems","volume":"9","author":"Vincent","year":"2015","journal-title":"IEEE Trans. Biomed. Circuits Syst."},{"key":"ncead49cebib20","doi-asserted-by":"publisher","DOI":"10.1063\/5.0038521","article-title":"Domain wall-magnetic tunnel junction spin\u2013orbit torque devices and circuits for in-memory computing","volume":"118","author":"Alamdar","year":"2021","journal-title":"Appl. Phys. Lett."},{"key":"ncead49cebib21","doi-asserted-by":"publisher","DOI":"10.1063\/1.2976435","article-title":"Tunnel magnetoresistance of 604% at 300K by suppression of Ta diffusion in CoFeB\/ MgO\/ CoFeB pseudo-spin-valves annealed at high temperature","volume":"93","author":"Ikeda","year":"2008","journal-title":"Appl. Phys. Lett."},{"key":"ncead49cebib22","doi-asserted-by":"publisher","first-page":"25","DOI":"10.1038\/ncomms1024","article-title":"Enhanced stochasticity of domain wall motion in magnetic racetracks due to dynamic pinning","volume":"1","author":"Jiang","year":"2010","journal-title":"Nat. Commun."},{"key":"ncead49cebib23","doi-asserted-by":"publisher","first-page":"1658","DOI":"10.1109\/TED.2021.3111846","article-title":"Voltage-controlled energy-efficient domain wall synapses with stochastic distribution of quantized weights in the presence of thermal noise and edge roughness","volume":"69","author":"Al Misba","year":"2021","journal-title":"IEEE Trans. Electron Devices"},{"key":"ncead49cebib24","doi-asserted-by":"publisher","first-page":"84946","DOI":"10.1109\/ACCESS.2022.3196688","article-title":"Energy efficient learning with low resolution stochastic domain wall synapse for deep neural networks","volume":"10","author":"Al Misba","year":"2022","journal-title":"IEEE Access"},{"key":"ncead49cebib25","doi-asserted-by":"publisher","first-page":"246","DOI":"10.1038\/s41928-018-0054-8","article-title":"Mixed-precision in-memory computing","volume":"1","author":"Le Gallo","year":"2018","journal-title":"Nat. Electron."},{"key":"ncead49cebib26","doi-asserted-by":"publisher","first-page":"406","DOI":"10.3389\/fnins.2020.00406","article-title":"Mixed-precision deep learning based on computational memory","volume":"14","author":"Nandakumar","year":"2020","journal-title":"Front. Neurosci."},{"key":"ncead49cebib27","first-page":"1","article-title":"A detailed analysis of the KDD CUP 99 data set","author":"Tavallaee","year":"2009"},{"key":"ncead49cebib28","doi-asserted-by":"publisher","first-page":"4396","DOI":"10.3390\/app9204396","article-title":"Machine learning and deep learning methods for intrusion detection systems: a survey","volume":"9","author":"Liu","year":"2019","journal-title":"Appl. Sci."},{"key":"ncead49cebib29","doi-asserted-by":"publisher","first-page":"578","DOI":"10.1016\/j.cose.2018.05.010","article-title":"Early-stage malware prediction using recurrent neural networks","volume":"77","author":"Rhode","year":"2018","journal-title":"Comput. Secur."},{"key":"ncead49cebib30","doi-asserted-by":"publisher","first-page":"708","DOI":"10.1016\/j.procs.2015.08.220","article-title":"Survey on anomaly detection using data mining techniques","volume":"60","author":"Agrawal","year":"2015","journal-title":"Proc. Comput. Sci."},{"key":"ncead49cebib31","doi-asserted-by":"publisher","first-page":"484","DOI":"10.1016\/j.ins.2016.04.019","article-title":"Fuzziness based semi-supervised learning approach for intrusion detection system","volume":"378","author":"Ashfaq","year":"2017","journal-title":"Inf. Sci."},{"key":"ncead49cebib32","first-page":"92","article-title":"Performance analysis of NSL-KDD dataset using ANN","author":"Ingre","year":"2015"},{"key":"ncead49cebib33","doi-asserted-by":"publisher","first-page":"e4150","DOI":"10.1002\/ett.4150","article-title":"Network intrusion detection system: a systematic study of machine learning and deep learning approaches","volume":"32","author":"Ahmad","year":"2021","journal-title":"Trans. Emerg. Telecommun. Technol."},{"key":"ncead49cebib34","first-page":"305","article-title":"Outside the closed world: on using machine learning for network intrusion detection","author":"Sommer","year":"2010"},{"key":"ncead49cebib35","first-page":"1","article-title":"Memristor based autoencoder for unsupervised real-time network intrusion and anomaly detection","author":"Alam","year":"2019"},{"key":"ncead49cebib36","doi-asserted-by":"publisher","DOI":"10.1063\/1.4939446","article-title":"Ultra-low switching energy and scaling in electric-field-controlled nanoscale magnetic tunnel junctions with high resistance-area product","volume":"108","author":"Grezes","year":"2016","journal-title":"Appl. Phys. Lett."},{"key":"ncead49cebib37","doi-asserted-by":"publisher","DOI":"10.1038\/srep31272","article-title":"Voltage controlled core reversal of fixed magnetic skyrmions without a magnetic field","volume":"6","author":"Bhattacharya","year":"2016","journal-title":"Sci. Rep."},{"key":"ncead49cebib38","doi-asserted-by":"publisher","first-page":"539","DOI":"10.1038\/s41928-020-0432-x","article-title":"Creation and annihilation of non-volatile fixed magnetic skyrmions using voltage control of magnetic anisotropy","volume":"3","author":"Bhattacharya","year":"2020","journal-title":"Nat. Electron."},{"key":"ncead49cebib39","doi-asserted-by":"publisher","DOI":"10.1063\/1.4923350","article-title":"Strain-mediated 180 perpendicular magnetization switching of a single domain multiferroic structure","volume":"118","author":"Li","year":"2015","journal-title":"J. Appl. Phys."},{"key":"ncead49cebib40","doi-asserted-by":"publisher","DOI":"10.1063\/1.3624900","article-title":"Hybrid spintronics and straintronics: a magnetic technology for ultra low energy computing and signal processing","volume":"99","author":"Roy","year":"2011","journal-title":"Appl. Phys. Lett."},{"key":"ncead49cebib41","doi-asserted-by":"publisher","first-page":"1378","DOI":"10.1038\/ncomms2386","article-title":"Strain-controlled magnetic domain wall propagation in hybrid piezoelectric\/ferromagnetic structures","volume":"4","author":"Lei","year":"2013","journal-title":"Nat. Commun."},{"key":"ncead49cebib42","doi-asserted-by":"publisher","DOI":"10.1063\/1.4893617","article-title":"Complete magnetization reversal in a magnetostrictive nanomagnet with voltage-generated stress: a reliable energy-efficient non-volatile magneto-elastic memory","volume":"105","author":"Biswas","year":"2014","journal-title":"Appl. Phys. Lett."},{"key":"ncead49cebib43","doi-asserted-by":"publisher","first-page":"5681","DOI":"10.1021\/acs.nanolett.6b02342","article-title":"Acoustic-wave-induced magnetization switching of magnetostrictive nanomagnets from single-domain to nonvolatile vortex states","volume":"16","author":"Sampath","year":"2016","journal-title":"Nano Lett."},{"key":"ncead49cebib44","doi-asserted-by":"publisher","first-page":"L1","DOI":"10.1016\/0304-8853(96)00062-5","article-title":"Current-driven excitation of magnetic multilayers","volume":"159","author":"Slonczewski","year":"1996","journal-title":"J. Magn. Magn. Mater."},{"key":"ncead49cebib45","doi-asserted-by":"publisher","first-page":"3910","DOI":"10.1038\/ncomms4910","article-title":"Chiral spin torque arising from proximity-induced magnetization","volume":"5","author":"Ryu","year":"2014","journal-title":"Nat. Commun."},{"key":"ncead49cebib46","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevApplied.14.014088","article-title":"Acoustic-wave-induced ferromagnetic-resonance-assisted spin-torque switching of perpendicular magnetic tunnel junctions with anisotropy variation","volume":"14","author":"Al Misba","year":"2020","journal-title":"Phys. Rev. Appl."},{"key":"ncead49cebib47","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6528\/ab6234","article-title":"Voltage control of domain walls in magnetic nanowires for energy-efficient neuromorphic devices","volume":"31","author":"Azam","year":"2020","journal-title":"Nanotechnology"},{"key":"ncead49cebib48","doi-asserted-by":"publisher","DOI":"10.1002\/aelm.202200563","article-title":"Shape\u2010dependent multi\u2010weight magnetic artificial synapses for neuromorphic computing","volume":"8","author":"Leonard","year":"2022","journal-title":"Adv. Electron. Mater."},{"key":"ncead49cebib49","article-title":"Binaryconnect: training deep neural networks with binary weights during propagations","author":"Courbariaux","year":"2015"},{"key":"ncead49cebib50","doi-asserted-by":"publisher","first-page":"46","DOI":"10.1016\/j.procs.2015.09.145","article-title":"Using data mining algorithms for developing a model for intrusion detection system (IDS)","volume":"61","author":"Duque","year":"2015","journal-title":"Proc. Comput. Sci."},{"key":"ncead49cebib51","first-page":"2704","article-title":"Quantization and training of neural networks for efficient integer-arithmetic-only inference","author":"Jacob","year":"2018"},{"key":"ncead49cebib52","first-page":"873","article-title":"High speed toggle MRAM with MgO-based tunnel junctions","author":"Slaughter","year":"2005"},{"key":"ncead49cebib53","doi-asserted-by":"publisher","DOI":"10.1063\/1.4899186","article-title":"The design and verification of MuMax3","volume":"4","author":"Vansteenkiste","year":"2014","journal-title":"AIP Adv."},{"key":"ncead49cebib54","doi-asserted-by":"publisher","DOI":"10.1063\/5.0046032","article-title":"A domain wall-magnetic tunnel junction artificial synapse with notched geometry for accurate and efficient training of deep neural networks","volume":"118","author":"Liu","year":"2021","journal-title":"Appl. Phys. Lett."},{"key":"ncead49cebib55","doi-asserted-by":"crossref","DOI":"10.1109\/TNNLS.2024.3369969","article-title":"Quantized magnetic domain wall synapse for efficient deep neural networks","author":"Dhull","year":"2024"},{"key":"ncead49cebib56","doi-asserted-by":"publisher","first-page":"2549","DOI":"10.1038\/s41467-021-22768-y","article-title":"Synaptic metaplasticity in binarized neural networks","volume":"12","author":"Laborieux","year":"2021","journal-title":"Nat. Commun."},{"key":"ncead49cebib57","doi-asserted-by":"publisher","first-page":"86","DOI":"10.1109\/JETCAS.2018.2796379","article-title":"Multiscale co-design analysis of energy, latency, area, and accuracy of a ReRAM analog neural training accelerator","volume":"8","author":"Marinella","year":"2018","journal-title":"IEEE J. Emerg. Sel. Top. Circuits Syst."}],"container-title":["Neuromorphic Computing and Engineering"],"original-title":[],"link":[{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2634-4386\/ad49ce","content-type":"text\/html","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2634-4386\/ad49ce\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2634-4386\/ad49ce","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2634-4386\/ad49ce\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2634-4386\/ad49ce\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2634-4386\/ad49ce\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2634-4386\/ad49ce\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"similarity-checking"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2634-4386\/ad49ce\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,6,20]],"date-time":"2024-06-20T11:18:24Z","timestamp":1718882304000},"score":1,"resource":{"primary":{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2634-4386\/ad49ce"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,6,1]]},"references-count":57,"journal-issue":{"issue":"2","published-online":{"date-parts":[[2024,6,20]]},"published-print":{"date-parts":[[2024,6,1]]}},"URL":"https:\/\/doi.org\/10.1088\/2634-4386\/ad49ce","relation":{},"ISSN":["2634-4386"],"issn-type":[{"value":"2634-4386","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,6,1]]},"assertion":[{"value":"Quantized non-volatile nanomagnetic domain wall synapse based autoencoder for efficient unsupervised network anomaly detection","name":"article_title","label":"Article Title"},{"value":"Neuromorphic Computing and Engineering","name":"journal_title","label":"Journal Title"},{"value":"paper","name":"article_type","label":"Article Type"},{"value":"\u00a9 2024 The Author(s). Published by IOP Publishing Ltd","name":"copyright_information","label":"Copyright Information"},{"value":"2024-01-16","name":"date_received","label":"Date Received","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2024-05-10","name":"date_accepted","label":"Date Accepted","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2024-06-20","name":"date_epub","label":"Online publication date","group":{"name":"publication_dates","label":"Publication dates"}}]}}