[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-120322-en":3,"doc-seo-120322-105":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},120322,1099514067415,"Rowan","https://ap-avatar.wpscdn.com/avatar/100002539d78ffe74a7?x-image-process=image/resize,m_fixed,w_180,h_180&k=1779092875211072502",8,"Research & Report","Quantum Machine Learning Architecture Search via Deep Reinforcement Learning","The rapid progress of quantum computing and machine learning has enabled quantum machine learning (QML), which leverages quantum capabilities to advance learning models. Designing effective QML architectures remains challenging because high model expressivity often requires deep circuits that can be hard to execute on noisy intermediate-scale quantum (NISQ) hardware. This work applies deep reinforcement learning to search for QML model architectures for supervised classification without fixing an initial ansatz. The method adaptively updates learning objectives during training and, through numerical simulations, identifies variational quantum circuit designs achieving high accuracy with reduced gate depth, supporting efficient AI-driven quantum circuit design in the NISQ era.","Quantum Machine Learning Architecture Search via Deep Reinforcement Learning  \nXin Dai 1 , Tzu-Chieh Wei2 , Shinjae Yoo 1 , Samuel Yen-Chi Chen 1  \n1 Computational Science Initiative, Brookhaven National Laboratory  \n2 C.N. Yang Institute for Theoretical Physics and Department of Physics and Astronomy, Stony Brook University {xdai, [sjyoo](sjyoo}@bnl.gov)[}](sjyoo}@bnl.gov)[@bnl.gov](sjyoo}@bnl.gov) , [tzu-chieh.wei@stonybrook.edu](tzu-chieh.wei@stonybrook.edu) , [ycchen1989@ieee.org](ycchen1989@ieee.org)  \narXiv :2407 .20147v1 [ quant-ph] 29 Jul 2024  \nAbstract—The rapid advancement of quantum computing (QC) and machine learning (ML) has given rise to the burgeoning field of quantum machine learning (QML), aiming to capitalize on the strengths of quantum computing to propel ML forward. Despite its promise, crafting effective QML models necessitates profound expertise to strike a delicate balance between model intricacy and feasibility on Noisy Intermediate-Scale Quantum (NISQ) devices. While complex models offer robust representation capabilities, their extensive circuit depth may impede seamless execution on extant noisy quantum platforms. In this paper, we address this quandary of QML model design by employing deep reinforcement learning to explore proficient QML model architectures tailored for designated supervised learning tasks. Specifically, our methodology involves training an RL agent to devise policies that facilitate the discovery of QML models without predetermined ansatz. Furthermore, we integrate an adaptive mechanism to dynamically adjust the learning objectives, fostering continuous improvement in the agent’s learning process. Through extensive numerical simulations, we illustrate the efficacy of our approach within the realm of classification tasks. Our proposed method successfully identifies VQC architectures capable of achieving high classification accuracy while minimizing gate depth. This pioneering approach not only advances the study of AI-driven quantum circuit design but also holds significant promise for enhancing performance in the NISQ era.  \nIndex Terms—quantum machine learning, quantum neural networks, variational quantum circuits, quantum architecture search  \nI. INTRODUCTION  \nQuantum computing (QC) holds the potential to revolutionize computational tasks, offering distinct advantages over classical computers [1] . The convergence of advancementsin quantum hardware and machine learning applications has sparked a growing interest in exploring the synergies between these cutting-edge technologies. Although existing quantum computers still suffer from noise, a promising solution lies ina hybrid quantum-classical framework. Here, computational tasks are divided into two parts: one executed on a quantum computer and the other on a classical computer [2],[3] . Central to this paradigm is the Variational Quantum Algorithm (VQA)  \n[2], [3], which serves as the cornerstone of hybrid computing.  \nThis work was supported by the U.S. DOE, Office of Science, Office of High Energy Physics under award DE-SC-0012704 . This research used resources of the NERSC, under Contract No.DE-AC02-05CH11231 using NERSC award HEP-ERCAP0023403 .  \nQuantum machine learning (QML) algorithms heavily rely on VQAs, utilizing variational quantum circuits (VQCs) as trainable components akin to classical neural networks. QML has demonstrated remarkable success across various domains, including classification [4]–[8], time-series modeling [9], natural language processing [10]–[13], generative modeling [14]–[16], and reinforcement learning [17]–[24] . While existing QML models have shown promise, they often require expert knowledge to design effective quantum circuit architectures. For instance, the configuration of encoding and variational subcircuits within VQCs significantly influences model performance and the realization of potential quantum advantages [25] . Moreover, the vast search space of VQCs presents a challenge, given th","cbCaivFdZqKBcoXs","https://ap.wps.com/l/cbCaivFdZqKBcoXs","pdf",1498507,1,10,"English","en",105,"# Introduction\n## Background: Hybrid quantum-classical variational algorithms\n## Motivation: Limited expert knowledge and large circuit search space\n# Relevant Works\n## Quantum architecture search in quantum computing\n# Method Overview\n## Deep reinforcement learning with adaptive search of learning targets (RL-QMLAS)\n# QAS Problem and RL Formulation\n# Variational Quantum Circuit Target\n# Simulations and Results","[{\"question\":\"What problem does the paper address in quantum machine learning architecture design?\",\"answer\":\"It targets the difficulty of building QML models that balance strong representation power with feasible execution on noisy NISQ devices, where deep circuits can hinder performance.\"},{\"question\":\"How does RL-QMLAS search for QML architectures?\",\"answer\":\"An RL agent is trained to generate policies that discover suitable variational quantum circuit architectures without requiring a predetermined ansatz.\"},{\"question\":\"What improvements does the adaptive learning target provide?\",\"answer\":\"The adaptive mechanism dynamically adjusts the learning objectives during reinforcement learning, improving agent learning efficiency and reducing reliance on a fixed high pre-defined learning target.\"}]","Quantum Machine Learning Architecture Search via Deep Reinforcement Learning | PDF",1785729450,25,{"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},"quantum-machine-learning-architecture-search-via-deep-reinforcement-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/quantum-machine-learning-architecture-search-via-deep-reinforcement-learning/120322/",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 the paper address in quantum machine learning architecture design?","Question",{"text":75,"@type":76},"It targets the difficulty of building QML models that balance strong representation power with feasible execution on noisy NISQ devices, where deep circuits can hinder performance.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does RL-QMLAS search for QML architectures?",{"text":80,"@type":76},"An RL agent is trained to generate policies that discover suitable variational quantum circuit architectures without requiring a predetermined ansatz.",{"name":82,"@type":73,"acceptedAnswer":83},"What improvements does the adaptive learning target provide?",{"text":84,"@type":76},"The adaptive mechanism dynamically adjusts the learning objectives during reinforcement learning, improving agent learning efficiency and reducing reliance on a fixed high pre-defined learning target.","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},[93,97,101,105,110,115,120,123,128,131,134],{"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":21,"doc_module":4,"doc_module_name":46,"category_name":132,"show_sort_weight":21,"slug":133},"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":46,"category_name":136,"show_sort_weight":106,"slug":137},19,"General","general"]