[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-seo-290755-105":3,"detail-sidebar-cat-0-en-105":81,"doc-detail-290755-en":131},{"code":4,"msg":5,"data":6},0,"ok",{"site_id":7,"language":8,"slug":9,"title":10,"keywords":11,"description":12,"schema_data":13,"social_meta":74,"head_meta":76,"extra_data":78,"updated_unix":80},105,"en","superconducting-hyperdimensional-associative-memory-circuit-for-scalable-machine-learning-read-online-free","Superconducting Hyperdimensional Associative Memory Circuit for Scalable Machine Learning - read online free","","We propose a generalized architecture for a rapid-single-flux-quantum (RSFQ) associative memory circuit that leverages hyperdimensional computing (HDC) for machine learning. The approach uses thousands-dimensional vectors and targets scalable SFQ ML by reducing memory footprint, simplifying computations, and using simpler training than superconducting neural-network accelerators. The superconducting HDC design runs fully on-chip at 33.3 GHz, integrates memory with logic, supports general ML tasks, and is manufacturable under current SFQ fabrication limits.",{"@graph":14,"@context":73},[15,34,56],{"@type":16,"itemListElement":17},"BreadcrumbList",[18,23,27,31],{"item":19,"name":20,"@type":21,"position":22},"https://docshare.wps.com","Home","ListItem",1,{"item":24,"name":25,"@type":21,"position":26},"https://docshare.wps.com/document/","Document",2,{"item":28,"name":29,"@type":21,"position":30},"https://docshare.wps.com/document/research-report/","Research & Report",3,{"item":32,"name":10,"@type":21,"position":33},"https://docshare.wps.com/document/superconducting-hyperdimensional-associative-memory-circuit-for-scalable-machine-learning-read-online-free/290755/",4,{"url":32,"name":10,"@type":35,"image":36,"author":41,"headline":10,"publisher":44,"fileFormat":47,"inLanguage":8,"description":12,"dateModified":48,"datePublished":49,"encodingFormat":47,"isAccessibleForFree":50,"interactionStatistic":51},"DigitalDocument",{"url":37,"@type":38,"width":39,"height":40},"https://docshare.wps.com/thumbnails/superconducting-hyperdimensional-associative-memory-circuit-for-scalable-machine-learning-read-online-free/290755.png","ImageObject",300,407,{"name":42,"@type":43},"Riley","Person",{"url":19,"name":45,"@type":46},"DocShare","Organization","application/pdf","2026-11-04","2026-09-17",true,{"@type":52,"interactionType":53,"userInteractionCount":55},"InteractionCounter",{"@type":54},"ViewAction",16,{"@type":57,"mainEntity":58},"FAQPage",[59,65,69],{"name":60,"@type":61,"acceptedAnswer":62},"What circuit architecture is proposed in the document?","Question",{"text":63,"@type":64},"It proposes a generalized architecture for a rapid-single-flux-quantum (RSFQ) associative memory circuit using hyperdimensional computing for machine learning.","Answer",{"name":66,"@type":61,"acceptedAnswer":67},"Why is hyperdimensional computing considered suitable for scalable RSFQ machine learning?",{"text":68,"@type":64},"HDC can use small memory footprints, simpler computations, and simpler training algorithms than superconducting neural network accelerators, supporting scalable SFQ ML.",{"name":70,"@type":61,"acceptedAnswer":71},"What performance and applicability claims are made about the superconducting HDC (SHDC) circuit?",{"text":72,"@type":64},"The SHDC circuit is described as operating at 33.3 GHz, fully on-chip with RSFQ memory integrated with logic, applicable to general ML tasks, and manufacturable at practically useful scales, including online learning for real-world ML tasks.","https://schema.org",{"og:url":32,"og:type":75,"og:title":10,"og:site_name":45,"og:description":12},"article",{"robots":77,"canonical":32},"index,follow",{"doc_id":79,"site_id":7},290755,1789642508,{"code":4,"msg":82,"data":83},"success",[84,88,92,96,101,106,111,115,120,123,127],{"id":22,"doc_module":4,"doc_module_name":25,"category_name":85,"show_sort_weight":86,"slug":87},"Story & Novel",90,"story-novel",{"id":26,"doc_module":4,"doc_module_name":25,"category_name":89,"show_sort_weight":90,"slug":91},"Literature",80,"literature",{"id":33,"doc_module":4,"doc_module_name":25,"category_name":93,"show_sort_weight":94,"slug":95},"Exam",70,"exam",{"id":97,"doc_module":4,"doc_module_name":25,"category_name":98,"show_sort_weight":99,"slug":100},5,"Comic",60,"comic",{"id":102,"doc_module":4,"doc_module_name":25,"category_name":103,"show_sort_weight":104,"slug":105},6,"Technology",50,"technology",{"id":107,"doc_module":4,"doc_module_name":25,"category_name":108,"show_sort_weight":109,"slug":110},7,"Healthcare",40,"healthcare",{"id":112,"doc_module":4,"doc_module_name":25,"category_name":29,"show_sort_weight":113,"slug":114},8,30,"research-report",{"id":116,"doc_module":4,"doc_module_name":25,"category_name":117,"show_sort_weight":118,"slug":119},9,"Religion & Spirituality",20,"religion-spirituality",{"id":118,"doc_module":4,"doc_module_name":25,"category_name":121,"show_sort_weight":118,"slug":122},"World Cup","world-cup",{"id":124,"doc_module":4,"doc_module_name":25,"category_name":125,"show_sort_weight":124,"slug":126},10,"Lifestyle","lifestyle",{"id":128,"doc_module":4,"doc_module_name":25,"category_name":129,"show_sort_weight":97,"slug":130},19,"General","general",{"code":4,"msg":82,"data":132},{"doc_id":79,"user_id":133,"nickname":42,"user_avatar":134,"doc_module":4,"category_id":112,"category_name":29,"doc_title":10,"doc_description":12,"doc_content":135,"file_id":136,"file_url":137,"file_type":138,"file_size":139,"view_count":55,"is_deleted":4,"is_public":22,"is_downloadable":22,"audit_status":22,"page_count":55,"language":140,"language_code":8,"site_id":7,"html_lang":8,"table_of_contents":141,"faqs":142,"seo_title":143,"seo_description":12,"update_tm":80,"read_time":109,"preview_page_count":4},1374391975076,"https://ap-avatar.wpscdn.com/avatar/14000253ca4ec9f6853?x-image-process=image/resize,m_fixed,w_180,h_180&k=1783305029341752051","Lawrence Berkeley National Laboratory  \nLBL Publications  \nTitle  \nSuperconducting Hyperdimensional Associative Memory Circuit for Scalable Machine Learning  \nPermalink  \n[https://escholarship.org/uc/item/2d0360xb](https://escholarship.org/uc/item/2d0360xb)  \nJournal  \nIEEE Transactions on Applied Superconductivity, 33(5)  \nISSN  \n1051-8223  \nAuthors  \nHuch, Kylie  \nGonzalez-Guerrero, Patricia Lyles, Darren  \net al.  \nPublication Date  \n2023  \nDOI  \n10.1109/tasc.2023.3271951  \nCopyright Information  \nThis work is made available under the terms of a Creative Commons Attribution License, available at [https://creativecommons.org/licenses/by/4.0/](https://creativecommons.org/licenses/by/4.0/)  \nPeer reviewed  \n[eScholarship.org](eScholarship.org) Powered by the California Digital Library  \nUniversity of California  \nSuperconducting Hyperdimensional Associative Memory Circuit for Scalable Machine Learning  \nKylie Huch, Patricia Gonzalez-Guerrero, Member, IEEE, Darren Lyles, Member, IEEE, and George  \nMichelogiannakis, Senior Member, IEEE  \nAbstract—We propose a generalized architecture for the first rapid-single-flux-quantum ( RSFQ) a ssociative m emory circuit. The circuit employs hyperdimensional computing (HDC), a machine learning (ML) paradigm utilizing vectors with dimensionality in the thousands to represent information. HDC designs have small memory footprints, simple computations, and simple training algorithms compared to superconducting neural network accelerators (SNNAs), making them a better option for scalable SFQ machine learning (ML) solutions. The proposed superconducting HDC (SHDC) circuit uses entirely on-chip RSFQ memory which is tightly integrated with logic, operates at 33.3 GHz, is applicable to general ML tasks, and is manufacturable at practically useful scales given current SFQ fabrication limits. Tailored to a language recognition task, SHDC consists of ∼2-20M Josephson junctions (JJs) and is 102 (RSFQ) to > 103 (ERSFQ) times more energy efficient than an analogous CMOS HDC circuit including cooling and > 104 (RSFQ) to 106 (ERSFQ) times excluding cooling while achieving 78-84% higher throughput. SHDC is up to 107 (RSFQ) to 106 (ERSFQ) times more energy efficient t han t he s tate o ft he a rt R SFQ SNNA, SuperNPU, while achieving 62-99% higher throughput for all but the smallest NN accelerated by SuperNPU. To the best of the authors’ knowledge, SHDC is currently the only superconducting ML approach feasible at practically useful scales for real-world ML tasks and capable of online learning.  \nI. INTRODUCTION  \nWITH the slowdown of Moore’s law and the end of  \nDennard scaling, superconducting digital computing offers a promising alternative for future high performance computing (HPC) systems due to its ability to operate at up to ∼ 100 GHz with low power dissipation [1], [2] . Of the variety of different superconducting digital logic families based on single-flux-quantum (SFQ) pulses, rapid-single-fluxquantum (RSFQ) logic is the most mature and remains the most common for high-speed circuit applications [3] . As such, we design and simulate a practical HDC circuit based on standard RSFQ logic gates [4] .  \nLimited device density is one of the greatest challenges for superconducting digital computing currently, making area a critical constraint for SFQ circuits [1], [5] . Although RSFQ circuits with about one million Josephson junctions (JJs) [6],[7] and, more recently, close to ten million JJs [8] have  \nThe authors are with the Computer Architecture Group, Applied Mathematics and Computational Research Division, Lawrence Berkeley National Laboratory, Berkeley, California, USA., Email:[kyliehuch,lg4er,dlyles,mihelog]@[lbl.gov](lbl.gov)  \nbeen demonstrated, these chips had highly-regular shift register designs. In terms of complex logic and irregular RSFQ circuits, recently demonstrated chips have been limited to around 20-30 thousand JJs [9] . Additionally, cryogenic onchip memory is widely regarded as a s","cbCaihuYFKtZ4ncH","https://ap.wps.com/l/cbCaihuYFKtZ4ncH","pdf",1219652,"English","# Abstract\n# I. Introduction\n## A. Background and Related Work","[{\"question\":\"What circuit architecture is proposed in the document?\",\"answer\":\"It proposes a generalized architecture for a rapid-single-flux-quantum (RSFQ) associative memory circuit using hyperdimensional computing for machine learning.\"},{\"question\":\"Why is hyperdimensional computing considered suitable for scalable RSFQ machine learning?\",\"answer\":\"HDC can use small memory footprints, simpler computations, and simpler training algorithms than superconducting neural network accelerators, supporting scalable SFQ ML.\"},{\"question\":\"What performance and applicability claims are made about the superconducting HDC (SHDC) circuit?\",\"answer\":\"The SHDC circuit is described as operating at 33.3 GHz, fully on-chip with RSFQ memory integrated with logic, applicable to general ML tasks, and manufacturable at practically useful scales, including online learning for real-world ML tasks.\"}]","Superconducting Hyperdimensional Associative Memory Circuit for Scalable Machine Learning - read online free | PDF"]