[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-124299-en":3,"doc-seo-124299-105":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":4,"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},124299,1374391974468,"Eden","https://ap-avatar.wpscdn.com/davatar_29158cc5080c5b710cf443261637dec0",8,"Research & Report","Democratizing Machine Learning and Reasoning Systems with Brain-inspired Efficient AI","Yang Ni’s dissertation investigates brain-inspired, hyperdimensional computing approaches to make machine learning and reasoning systems more efficient and practical. The work focuses on algorithm-hardware co-design for edge deployment, including mapping learning computation to Edge TPU and improving training or inference efficiency. It further develops hyperdimensional reinforcement learning for continuous control, and introduces uncertainty quantification techniques for trustworthy hyperdimensional computing. Finally, it proposes brain-inspired efficient active learning using uncertainty estimation and ensemble efficiency, validated through extensive experimental evaluations.","UC Irvine  \nUC Irvine Electronic Theses and Dissertations  \nTitle  \nDemocratizing Machine Learning and Reasoning Systems with Brain-inspired Efficient AI  \nPermalink  \n[https://escholarship.org/uc/item/1nh0r9j4](https://escholarship.org/uc/item/1nh0r9j4)  \nAuthor  \nNi, Yang  \nPublication Date  \n2025  \nCopyright Information  \nThis work is made available under the terms of a Creative Commons AttributionNonCommercial-NoDerivatives License, available at [https://creativecommons.org/licenses/by-nc-nd/4.0/](https://creativecommons.org/licenses/by-nc-nd/4.0/)  \n[Peer reviewed|Thesis/dissertation](Peer reviewed|Thesis/dissertation)  \n[eScholarship.org](eScholarship.org) Powered by the California Digital Library  \nUniversity of California  \nUNIVERSITY OF CALIFORNIA,  \nIRVINE  \nDemocratizing Machine Learning and Reasoning Systems with Brain-inspired Efficient AI  \nDISSERTATION  \nsubmitted in partial satisfaction of the requirements for the degree of  \nDOCTOR OF PHILOSOPHY  \nin Computer Science  \nby  \nYang Ni  \nDissertation Committee:  \nAssistant Professor Mohsen Imani, Chair Professor Tony Givargis  \nAssistant Professor Sang-Woo Jun Assistant Professor Hamidreza Aghasi  \nChapter 2 © 2022 European Design and Automation Association (EDAA)  \nChapter 3 © 2022 Yang Ni  \nChapter 4 © 2023 IEEE  \nChapter 5 © 2024 Yang Ni  \nChapter 6 © 2024 Yang Ni All other materials © 2025 Yang Ni  \nDEDICATION  \nTo my wife, Han,  \nand my parents, Bangliang and Xiaoning for all their support, kindness, and love.  \nTABLE OF CONTENTS  \nPage  \nLIST OF FIGURES vi  \nLIST OF TABLES ix  \nLIST OF ALGORITHMS x  \nACKNOWLEDGMENTS xi  \nVITA xiii  \nABSTRACT OF THE DISSERTATION xvii  \n1 Introduction 1  \n1.1 Overview ...................................... 1  \n1.2 Background: Brain-inspired Hyperdimensional Computing .......... 4  \n2 Algorithm-Hardware Co-Design for Efficient Hyperdimensional Learning 7  \n2.1 Introduction .................................... 7  \n2.2 Related Work ................................... 11  \n2.3 Hyperdimensional Computing at the Edge ................... 12  \n2.3.1 Mapping the hyperdimensional computing to Edge TPU ....... 13  \n2.3.2 Boost the Efficiency with Bagging .................... 16  \n2.4 Experimental Results ............................... 19  \n2.4.1 Experimental setup ............................ 19  \n2.4.2 Training Efficiency ............................ 21  \n2.4.3 Inference Efficiency and Accuracy .................... 22  \n2.4.4 Parameter Search for Bagging ...................... 24  \n2.4.5 Encoding Scalability with Number of Input Feature .......... 24  \n2.5 Conclusion ..................................... 26  \n3 Hyperdimensional Reinforcement Learning for Continuous Control 27  \n3.1 Introduction .................................... 27  \n3.2 Preliminaries ................................... 31  \n3.2.1 Reinforcement Learning ......................... 31  \n3.3 Vanilla policy gradient method ......................... 32  \n3.4 Hyperdimensional Policy-based RL ....................... 34  \n3.4.1 Critic: State Value Approximation ................... 36  \n3.4.2 Actor: Policy Gradient with Clipping .................. 38  \n3.5 Experimental Result ............................... 39  \n3.5.1 Experiment Settings ........................... 39  \n3.5.2 Rewards and Runtime Comparison ................... 40  \n3.5.3 HDPG Efficiency on Different Platforms ................ 40  \n3.5.4 HDPG & Dimensionality ......................... 42  \n3.6 Related Work ................................... 42  \n3.7 Conclusion ..................................... 44  \n4 Trustworthy Hyperdimensional Computing with Uncertainty Quantification 45  \n4.1 Introduction .................................... 45  \n4.2 Related Work ................................... 48  \n4.3 DiceHD:nabling Efficient Bayesian HDC ..................... 49  \n4.3.1 VFA & Hyperdimensional Regression .................. 50  \n4.3.2 Hyperdimensional Regression with Uncertainty Estimation ...... 52  \n4.3.3 DiceHD P","cbCaidKV1YxqpTnR","https://ap.wps.com/l/cbCaidKV1YxqpTnR","pdf",10006723,1,171,"English","en",105,"# Introduction\n## Overview\n## Background: Brain-inspired Hyperdimensional Computing\n# Algorithm-Hardware Co-Design for Efficient Hyperdimensional Learning\n## Introduction\n## Related Work\n## Hyperdimensional Computing at the Edge\n## Experimental Results\n## Conclusion\n# Hyperdimensional Reinforcement Learning for Continuous Control\n## Introduction\n## Preliminaries\n## Vanilla policy gradient method\n## Hyperdimensional Policy-based RL\n## Experimental Result\n## Related Work\n## Conclusion\n# Trustworthy Hyperdimensional Computing with Uncertainty Quantification\n## Introduction\n## Related Work\n## DiceHD:Enabling Efficient Bayesian HDC\n## Experimental Results\n## Conclusion\n# Brain-inspired Hyperdimensional Efficient Active Learning\n## Introduction\n## Background & Related Works\n## Estimating Uncertainty in HDC via Efficient Ensemble\n## HDC-based Active Learning\n## Experiments\n## Conclusion","[{\"question\":\"What problem does the dissertation address for machine learning and reasoning systems?\",\"answer\":\"It targets efficiency and practicality by leveraging brain-inspired hyperdimensional computing for scalable learning and reasoning, especially in edge settings and trustworthy scenarios.\"},{\"question\":\"How does the work improve hyperdimensional learning efficiency on edge hardware?\",\"answer\":\"It presents algorithm-hardware co-design, including mapping hyperdimensional computing to Edge TPU and using efficiency-enhancing strategies such as bagging to boost performance.\"},{\"question\":\"How are uncertainty and trust handled in the hyperdimensional computing approach?\",\"answer\":\"The dissertation introduces uncertainty quantification methods, including an approach labeled DiceHD, which performs hyperdimensional regression with uncertainty estimation and evaluates runtime and energy efficiency.\"}]","Democratizing Machine Learning and Reasoning Systems with Brain-inspired Efficient AI | PDF",1785821469,431,{"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},"democratizing-machine-learning-and-reasoning-systems-with-brain-inspired-efficient-ai","",{"@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/democratizing-machine-learning-and-reasoning-systems-with-brain-inspired-efficient-ai/124299/",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","2026-08-04",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What problem does the dissertation address for machine learning and reasoning systems?","Question",{"text":75,"@type":76},"It targets efficiency and practicality by leveraging brain-inspired hyperdimensional computing for scalable learning and reasoning, especially in edge settings and trustworthy scenarios.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does the work improve hyperdimensional learning efficiency on edge hardware?",{"text":80,"@type":76},"It presents algorithm-hardware co-design, including mapping hyperdimensional computing to Edge TPU and using efficiency-enhancing strategies such as bagging to boost performance.",{"name":82,"@type":73,"acceptedAnswer":83},"How are uncertainty and trust handled in the hyperdimensional computing approach?",{"text":84,"@type":76},"The dissertation introduces uncertainty quantification methods, including an approach labeled DiceHD, which performs hyperdimensional regression with uncertainty estimation and evaluates runtime and energy efficiency.","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"]