[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-121006-en":3,"doc-seo-121006-105":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},121006,4398048950312,"Violet","https://ap-avatar.wpscdn.com/avatar/400002538284de19e3c?_k=1778320343897328908",8,"Research & Report","Better than classical? - The subtle art of benchmarking quantum machine learning models","Benchmarking quantum machine learning models is difficult because experimental design, the limited problem scales accessible today, and commercial-driven narratives can distort conclusions. To support better decisions, the study develops an open-source benchmarking package built on the PennyLane framework. It evaluates 12 popular quantum machine learning models across 6 binary classification tasks, producing 160 datasets. Results show classical baselines typically outperform quantum classifiers, and removing entanglement often preserves or improves performance.","arXiv :2403 .07059v2 [ quant-ph] 14 Mar 2024  \nBetter than classical? The subtle art of benchmarking quantum machine learning  \nmodels  \nJoseph Bowles, 1, ∗ Shahnawaz Ahmed, 1, 2,† and Maria Schuld 1,‡  \n1 Xanadu, Toronto, ON, M5G 2C8, Canada  \n2 Chalmers University of Technology (Dated: March 15, 2024)  \nBenchmarking models via classical simulations is one of the main ways to judge ideas in quantum machine learning before noise-free hardware is available. However, the huge impact of the experimental design on the results, the small scales within reach today, as well as narratives influenced by the commercialisation of quantum technologies make it difficult to gain robust insights. To facilitate better decision-making we develop an open-source package based on the PennyLane software framework and use it to conduct a large-scale study that systematically tests 12 popular quantum machine learning models on 6 binary classification tasks used to create 160 individual datasets. We find that overall, out-of-the-box classical machine learning models outperform the quantum classifiers. Moreover, removing entanglement from a quantum model often results in as good or better performance, suggesting that “quantumness” may not be the crucial ingredient for the small learning tasks considered here. Our benchmarks also unlock investigations beyond simplistic leaderboard comparisons, and we identify five important questions for quantum model design that follow from our results.  \nMuch has been written about the “potential” of quantum machine learning, a discipline that asks how quantum computers fundamentally change what we can learn from data [1, 2] . While we have no means of running quantum algorithms on noise-free hardware yet, there are only a limited number of tools available to assess this potential. Besides proving advantages for artificial problem settings on paper, certain ideas – most prominently, variational models designed for near-term quantum technologies – can be tested in classical simulations using small datasets. Such benchmarks have in fact become a standard practice in the quantum machine learning literature and are found in almost every paper.  \nA taste for the results derived from small-scale benchmarks can be obtained through a simple literature review exercise. Out of 55 relevant papers published on the preprint server arXiv 1 until December 2023 that contain the terms “quantum machine learning” and “outperform” in title or abstract, one finds that about 40% claim that a quantum model outperforms a classical model, while about 50% claim that some improvement to a quantum machine learning method outperforms the original one (such as optimisers [3, 4], pre-training strategies [5] or symmetry-aware ansatze [6, 7]) . Only 3 papers or 4% find that a quantum model does not outperform a classi-  \n∗ [joseph@xanadu.ai](joseph@xanadu.ai)[ ](joseph@xanadu.ai)† [shahnawaz.ahmed95@gmail.com](shahnawaz.ahmed95@gmail.com)[ ](shahnawaz.ahmed95@gmail.com)‡ [maria@xanadu.ai](maria@xanadu.ai)  \n1 It is standard practice in the field of quantum computing to publish articles on the arXiv server, and we therefore expect it to provide representative samples ofthe literature. The 55 papers are a subset of 73 papers returned by the keyword search, and were selected by manually reading the abstracts and discarding papers that related to quantum-inspired methods, other fields than quantum machine learning, or did not explicitly mention a method outperforming another.  \nFIG. 1. The scope of the benchmark study at a glance.  \ncal one [8–10]; two of these are quick to mention that this is “due to the noise level in the available quantum hardware” [8] or that “the proposed methods are ready for [...] quantum computing devices which have the potential to outperform classical systems in the near future”  \n[10] . Only one paper [9] draws critical conclusions from their empirical results. If we assume that this literature review is representative2 , th","cbCaiia6StyTeGEG","https://ap.wps.com/l/cbCaiia6StyTeGEG","pdf",5845730,1,41,"English","en",105,"# Introduction\n## Motivation for evidence-based benchmarking\n# Benchmarking Methodology\n## Open-source package and experimental design\n## Model and dataset scope\n# Results and Implications\n## Out-of-the-box performance comparison\n## Entanglement ablation insights\n## Beyond leaderboard comparisons","[{\"question\":\"Why is benchmarking in near-term quantum machine learning challenging?\",\"answer\":\"Results can be heavily affected by experimental design choices, small achievable scales today, and narratives shaped by the commercialization of quantum technologies. These factors make robust insights harder to obtain.\"},{\"question\":\"How does the study benchmark quantum machine learning models?\",\"answer\":\"It builds an open-source package based on PennyLane and systematically tests 12 quantum machine learning models on 6 binary classification tasks. The process generates 160 individual datasets.\"},{\"question\":\"What do the benchmarks conclude about quantum classifiers versus classical models?\",\"answer\":\"Overall, out-of-the-box classical machine learning models outperform the quantum classifiers. Additionally, removing entanglement from a quantum model often yields performance that is as good or better for the small tasks considered.\"}]","Better than classical? - The subtle art of benchmarking quantum machine learning models | PDF",1785733278,103,{"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},"better-than-classical-the-subtle-art-of-benchmarking-quantum-machine-learning-models","",{"@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/better-than-classical-the-subtle-art-of-benchmarking-quantum-machine-learning-models/121006/",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":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"Why is benchmarking in near-term quantum machine learning challenging?","Question",{"text":75,"@type":76},"Results can be heavily affected by experimental design choices, small achievable scales today, and narratives shaped by the commercialization of quantum technologies. These factors make robust insights harder to obtain.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does the study benchmark quantum machine learning models?",{"text":80,"@type":76},"It builds an open-source package based on PennyLane and systematically tests 12 quantum machine learning models on 6 binary classification tasks. The process generates 160 individual datasets.",{"name":82,"@type":73,"acceptedAnswer":83},"What do the benchmarks conclude about quantum classifiers versus classical models?",{"text":84,"@type":76},"Overall, out-of-the-box classical machine learning models outperform the quantum classifiers. Additionally, removing entanglement from a quantum model often yields performance that is as good or better for the small tasks considered.","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"]