[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-119960-en":3,"doc-seo-119960-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},119960,4398048949847,"Eliana","https://ap-avatar.wpscdn.com/avatar/400002536579ef2da7f?_k=1778318612642679267",8,"Research & Report","TinyNS - Platform-Aware Neurosymbolic Auto Tiny Machine Learning","Edge machine learning enables intelligent, time-critical and remote inference, yet deploying interpretable AI that performs high-level symbolic reasoning and respects system rules and physics under tight platform resource constraints remains difficult. This paper introduces TinyNS, a platform-aware neurosymbolic architecture search framework for joint optimization of symbolic and neural operators. TinyNS automatically generates microcontroller code for neurosymbolic models, combines symbolic integrity with neural robustness, uses a gradient-free Bayesian optimizer across discontinuous and mixed spaces, and queries the target hardware during optimization. Experiments deploy microcontroller-class neurosymbolic models and show improved performance over purely neural or purely symbolic approaches while ensuring real-hardware execution.","UCLA  \nUCLA Previously Published Works  \nTitle  \nTinyNS: Platform-Aware Neurosymbolic Auto Tiny Machine Learning.  \nPermalink  \n[https://escholarship.org/uc/item/59n7w2rz](https://escholarship.org/uc/item/59n7w2rz)  \nJournal  \nACM Transactions on Embedded Computing Systems, 23(3)  \nAuthors  \nSaha, Swapnil  \nSandha, Sandeep Aggarwal, Mohit et al.  \nPublication Date  \n2024-05-01  \nDOI  \n10.1145/3603171  \nPeer reviewed  \n[eScholarship.org](eScholarship.org) Powered by the California Digital Library  \nUniversity of California  \nAuthor Manuscript Author Manuscript Author Manuscript Author Manuscript  \n\n|  | HHS Public Access\u003Cbr>Author manuscript\u003Cbr>ACM Trans Embed Comput Syst. Author manuscript; available in PMC 2024 June 26. |\n| --- | --- |\n\nPublished in final edited form as:  \nACM Trans Embed Comput Syst. 2024 May ; 23(3): . doi:10.1145/3603171 .  \nTinyNS: Platform-Aware Neurosymbolic Auto Tiny Machine Learning  \nSWAPNIL SAYAN SAHA,  \nUniversity of California-Los Angeles, Los Angeles, CA, USA SANDEEP SINGH SANDHA,  \nAbacus.AI, Seattle, WA, USA  \nMOHIT AGGARWAL,  \nBrightNight, Austin, TX, USA  \nBRIAN WANG,  \nUniversity of California-Los Angeles, Los Angeles, CA, USA  \nLIYING HAN,  \nUniversity of California-Los Angeles, Los Angeles, CA, USA  \nJULIAN DE GORTARI BRISENO,  \nUniversity of California-Los Angeles, Los Angeles, CA, USA  \nMANI SRIVASTAVA  \nUniversity of California-Los Angeles, Los Angeles, CA, USA  \nAbstract  \nMachine learning at the extreme edge has enabled a plethora of intelligent, time-critical, and remote applications. However, deploying interpretable artificial intelligence systems that can perform high-level symbolic reasoning and satisfy the underlying system rules and physics within the tight platform resource constraints is challenging. In this paper, we introduce TINYNS, the first platform-aware neurosymbolic architecture search framework for joint optimization of symbolic and neural operators. TINYNS provides recipes and parsers to automatically write microcontroller code for five types of neurosymbolic models, combining the context awareness and integrity of symbolic techniques with the robustness and performance of machine learning models. TINYNS uses a fast, gradient-free, black-box Bayesian optimizer over discontinuous, conditional, numeric, and categorical search spaces to find the best synergy of symbolic code and neural networks within the hardware resource budget. To guarantee deployability, TINYNS talks to the target hardware during the optimization process. We showcase the utility of TINYNS by deploying microcontroller-class neurosymbolic models through several case studies. In all use cases, TINYNS  \nPermission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than the author(s) must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions [from permissions@acm.org](from permissions@acm.org).  \nAuthors’ addresses: Swapnil Sayan Saha, [swapnilsayan@g.ucla.edu](swapnilsayan@g.ucla.edu).  \nAuthor Manuscript Author Manuscript Author Manuscript Author Manuscript  \nSAHA et al. Page 2  \noutperforms purely neural or purely symbolic approaches while guaranteeing execution on real hardware.  \nKeywords  \nneurosymbolic; neural architecture search; TinyML; AutoML; Bayesian; platform-aware  \n1 INTRODUCTION  \nTiny machine learning (TinyML) refers to hardware and software suites that enable alwayson, ultra-low-power (≤ 1 mW), and on-device sensor data analytics on low-end (≤ 1-2 MB of SRAM and eFlash) Internet of Things (IoT) platforms [51, 126, 136, 148] . TinyML holds the key to making on-board intellige","cbCaifXYJhc0US7O","https://ap.wps.com/l/cbCaifXYJhc0US7O","pdf",3609931,1,85,"English","en",105,"# Introduction\n## TinyML and its constraints\n## Neural architecture search and TinyML compilation\n## Motivation for neurosymbolic, platform-aware deployment","[{\"question\":\"What problem does TinyNS address in TinyML deployments?\",\"answer\":\"Deploying interpretable AI that can do symbolic reasoning while satisfying underlying system rules and physics under strict platform resource constraints is challenging. TinyNS targets this gap with a platform-aware neurosymbolic architecture search approach.\"},{\"question\":\"How does TinyNS search for the best symbolic-neural combination?\",\"answer\":\"TinyNS uses a fast gradient-free, black-box Bayesian optimizer over discontinuous and mixed search spaces. It jointly optimizes symbolic code and neural networks within the hardware resource budget.\"},{\"question\":\"How does TinyNS ensure models can run on real hardware?\",\"answer\":\"To guarantee deployability, TinyNS communicates with the target hardware during optimization. This allows the search to directly consider platform execution constraints, leading to working deployments on microcontroller-class systems.\"}]","TinyNS - Platform-Aware Neurosymbolic Auto Tiny Machine Learning | PDF",1785727228,214,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"tinyns-platform-aware-neurosymbolic-auto-tiny-machine-learning","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/tinyns-platform-aware-neurosymbolic-auto-tiny-machine-learning/119960/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-03",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What problem does TinyNS address in TinyML deployments?","Question",{"text":75,"@type":76},"Deploying interpretable AI that can do symbolic reasoning while satisfying underlying system rules and physics under strict platform resource constraints is challenging. TinyNS targets this gap with a platform-aware neurosymbolic architecture search approach.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does TinyNS search for the best symbolic-neural combination?",{"text":80,"@type":76},"TinyNS uses a fast gradient-free, black-box Bayesian optimizer over discontinuous and mixed search spaces. It jointly optimizes symbolic code and neural networks within the hardware resource budget.",{"name":82,"@type":73,"acceptedAnswer":83},"How does TinyNS ensure models can run on real hardware?",{"text":84,"@type":76},"To guarantee deployability, TinyNS communicates with the target hardware during optimization. This allows the search to directly consider platform execution constraints, leading to working deployments on microcontroller-class systems.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]