[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-117510-en":3,"doc-seo-117510-105":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},117510,1099514067438,"River Wang","https://ap-avatar.wpscdn.com/avatar/100002539ee87300030?x-image-process=image/resize,m_fixed,w_180,h_180&k=1780474512215547542",8,"Research & Report","Advances in Quantum Machine Learning - Doctoral Dissertation","This dissertation presents three advances in quantum machine learning. It introduces a variational hierarchical tensor-network method for classification and regression over both quantum and classical data. It also proposes an initialization strategy for parameterized quantum circuits to mitigate barren plateaus by avoiding random starts that yield exponentially small gradients. Finally, it models quantum systems interacting with unknown environments to capture non-Markovian effects. The work targets near-term constraints with small, shallow circuits under noise and evaluates behavior relevant to current quantum hardware.","Advances in Quantum Machine  \nLearning  \nEdward Grant  \nA dissertation submitted in partial fulfillment of the requirements for the degree of  \nDoctor of Philosophy  \nof  \nUniversity College London.  \nDepartment of Computer Science  \nUniversity College London  \nDecember 20, 2024  \n2  \nI, Edward Grant, confirm that the work presented in this thesis is my own. Where information has been derived from other sources, I confirm that this has been indicated in the work.  \nAbstract  \nThis dissertation thesis comprises three main contributions to the field of Quantum Machine Learning. The first is a variational machine learning method for classification and regression that employs a hierarchical structure based on tensor networks that can be trained to make predictions based on both quantum and classical data. The second is a method for initializing parameterized quantum circuits to avoid the problem of barren plateaus-a property of trainable quantum circuits where if the parameters of the circuit are initialized at random they can be effectively impossible to optimize because of gradients that become exponentially small in the number of qubits. The third contribution is a method for modelling quantum systems embedded in an unknown environment that can cause non-Markovian effects.  \nAcknowledgements  \nI would like thank my PhD Supervisors Simone Severini and Andrew Green.  \nIn addition I would like to thank collaborators, supporters and advisors, Leonard Wossnig, Marcello Benedetti, Shuxiang Cao, Hongxiang Chen, Jules Tilly, Dan Browne, Leonardo Banchi, Andrea Rocchetto, Paul Warbuton, Vid Stojevic, Lopa Murgai, Miriam Cha, Mateusz Ostaszewski, Ian Horrobin, Sergii Strelchuk, Giuseppe Carleo, Ivan Runger, Andrew Hallam, Joshua Lockhart, Michael Vasmer and Nikolas Breuckmann.  \nI would like to acknowledge the opportunity given to me to undertake this PhD by the University College London, Centre for Doctoral Training in Delivering Quantum Technologies and the financial support given by the Engineering and Physical Sciences Research Council (EPSRC) [EP/P510270/1] .  \nImpact statement  \nQuantum machine learning methods and parameterized quantum circuits can be used to model systems for which no efficient classical methods are known. Applications include the evaluation of observables in chemical systems where Hamiltonians are too complex to be efficiently evaluated by classical means. Additionally, quantum machine learning can be used to process and extract information directly from the wavefunction, which can defy efficient classical representation and processing.  \nThe techniques presented in this thesis contribute to an expanding field focused on utilizing quantum computers for learning from and modeling such systems. A key focus of this work is exploring how these methods might be employed in the presence of noise, which is prevalent in today's quantum computers. The approaches presented are tested primarily on small, shallow circuits, reflecting the current limitations of quantum hardware, and further exploration will be required to understand their behavior in larger, deeper circuits where noise becomes more significant.  \nContents  \n1 Introduction 15  \n1.1 Summary of contribution ....................... 20  \n1.2 Statement of authorship ........................ 21  \n2 Hierarchical quantum classifiers 24  \n2.1 Introduction .............................. 24  \n2.2 Data encoding ............................. 27  \n2.3 Circuit architecture .......................... 28  \n2.4 Unitary parameterization ....................... 29  \n2.5 Learning process and complexity ................... 32  \n2.6 Experimental results: Iris dataset ................... 33  \n2.6.1 Experimental results: Handwritten digits (MNIST) ..... 34  \n2.7 Experimental results: Quantum data ................. 36  \n2.8 Experimental results: Characterizing the effect of noise on classification performance .......................... 38  \n2.8.1 Experimental results: Deployment on a quantum compute","cbCainGFQsD8ocmg","https://ap.wps.com/l/cbCainGFQsD8ocmg","pdf",1174432,1,95,"English","en",105,"# Introduction\n## Summary of contribution\n## Statement of authorship\n# Hierarchical quantum classifiers\n## Data encoding\n## Circuit architecture\n## Unitary parameterization\n## Learning process and complexity\n## Experimental results: Iris dataset\n## Experimental results: Handwritten digits (MNIST)\n## Experimental results: Quantum data\n## Experimental results: Characterizing the effect of noise on classification performance\n## Discussion\n## Conclusion\n# Initializing and optimizing parameterized quantum circuits\n## A quick recap of the barren plateau problem\n## Initializing a circuit as a sequence of blocks of identity operators\n## Experimental results\n## Conclusion\n# Characterizing quantum circuit noise\n## Background\n## Non-Markovian processes\n## Recurrent Neural Networks\n## Main idea\n## Numerical experiments\n## Conclusion\n# General Conclusions","[{\"question\":\"What are the three main contributions of the dissertation in quantum machine learning?\",\"answer\":\"The work contributes: (1) a variational hierarchical tensor-network method for classification and regression using quantum and classical data; (2) a circuit initialization approach to avoid barren plateaus; and (3) a framework for modeling quantum systems with unknown environments that induce non-Markovian effects.\"},{\"question\":\"How does the dissertation address the barren plateau problem in parameterized quantum circuits?\",\"answer\":\"It proposes initializing circuits as structured sequences (blocks) of identity operators, aiming to prevent random parameter starts that make optimization ineffective by producing exponentially small gradients.\"},{\"question\":\"Why is quantum circuit noise a central focus, and how is it evaluated?\",\"answer\":\"Because noise is prevalent in current quantum computers, the approaches are tested mainly on small, shallow circuits to assess how noise impacts learning and classification performance, including experiments involving deployment on quantum hardware.\"}]","Advances in Quantum Machine Learning - Doctoral Dissertation | PDF",1785676475,239,{"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},"advances-in-quantum-machine-learning-doctoral-dissertation","",{"@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/advances-in-quantum-machine-learning-doctoral-dissertation/117510/",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","2026-08-02",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 are the three main contributions of the dissertation in quantum machine learning?","Question",{"text":75,"@type":76},"The work contributes: (1) a variational hierarchical tensor-network method for classification and regression using quantum and classical data; (2) a circuit initialization approach to avoid barren plateaus; and (3) a framework for modeling quantum systems with unknown environments that induce non-Markovian effects.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does the dissertation address the barren plateau problem in parameterized quantum circuits?",{"text":80,"@type":76},"It proposes initializing circuits as structured sequences (blocks) of identity operators, aiming to prevent random parameter starts that make optimization ineffective by producing exponentially small gradients.",{"name":82,"@type":73,"acceptedAnswer":83},"Why is quantum circuit noise a central focus, and how is it evaluated?",{"text":84,"@type":76},"Because noise is prevalent in current quantum computers, the approaches are tested mainly on small, shallow circuits to assess how noise impacts learning and classification performance, including experiments involving deployment on quantum hardware.","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"]