[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-119367-en":3,"doc-seo-119367-105":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},119367,1099514067438,"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","Inverse Polymer Design using Quantum Machine Learning - Master’s Thesis 2024","This thesis studies how quantum machine learning (QML) can address inverse polymer design, a goal-driven molecular generation task that produces polymers with targeted properties. The work centers on polyhydroxyalkanoates (PHAs), biodegradable polymers with notable industrial relevance. Objectives are twofold: develop simple quantum and classical models for generating PHAs and compare their performance, and evaluate whether QML provides an advantage over classical machine learning in this setting. Experiments train models on 1,510 PHA polymers, including quantum training under depolarizing noise, and assess test loss alongside validity, novelty, and uniqueness. Results indicate all models learn without overfitting; QML matches the simplest classical model at noise probability 10−3 or lower, while performance degrades substantially above 10−3.","Master’s Programme in Mathematics and Operations Research  \nInverse Polymer Design using Quantum Machine Learning  \nEino Yrjö-Koskinen  \nMaster’s Thesis 2024  \n© 2024  \nThis work is licensed under a Creative Commons“Attribution-NonCommercial-ShareAlike 4 .0 International” license.  \n| Author Eino Yrjö-Koskinen |\n| --- |\n| Title Inverse Polymer Design using Quantum Machine Learning |\n| Degree programme Mathematics and Operations Research |\n| Major Applied Mathematics |\n| Supervisor Asst. Prof. Alexandru Paler |\n| Advisor Dr. Olli Nurmi |\n| Collaborative partner VTT Technical Research Centre of Finland |\n| Date 18 Nov 2024 Number of pages 58 Language English |\n| Abstract\u003Cbr>This thesis explores the application of quantum machine learning (QML) to the inverse polymer design problem. It is a goal-oriented molecular generation task, where a machine learning (ML) model is used to generate polymers with desired properties. The focus is on generating polyhydroxyalkanoates (PHAs), a class of biodegradable polymers with significant industrial relevance.\u003Cbr>The research objectives are twofold: 1) to develop simple quantum and classical models capable of generating PHAs with desired properties and compare their performance, and 2) to assess whether QML offers an advantage over classical ML in the context of inverse polymer design.\u003Cbr>Computational experiments utilize a dataset of 1,510 PHA polymers to train multiple quantum and classical ML models, with a separate classical model trained to estimate the properties of the generated polymers. The quantum models are also trained using depolarizing noise with varying noise probabilities. Evaluation is conducted by comparing the test losses, as well as the validity, novelty, and uniqueness benchmarks. Results show that all implemented models generate valid polymers and successfully learn the relationship between polymers and their properties without overfitting.\u003Cbr>Notably, with a noise probability of 10−3 or lower, the most complex quantum model achieves performance comparable to the simplest classical model despite having significantly fewer trainable parameters. However, when the noise probability exceeds 10−3, none of the quantum models perform well.\u003Cbr>This study contributes to understanding QML’s capabilities and limitations in molecular design. Future work will focus on scaling and improving the models, addressing resource limitations, and extending evaluation to alternative data sets. |\n| Keywords Inverse polymer design, molecule generation, quantum machine learning, high performance computing |\n\nPreface  \nI would like to express my gratitude to Dr. Ville Kotovirta for providing me with the opportunity to work on this project and for his guidance during its early stages. Iam also thankful to Assistant Professor Alexandru Paler for his invaluable assistance throughout the writing process. Additionally, I want to thank to Dr. Olli Pakarinen, Choo Hyunkyung, and Jake Muff for their support. Finally, I wish to acknowledge CSC – IT Center for Science, Finland, for providing the computational resources.  \nOtaniemi, 18 November 2024  \nEino Yrjö-Koskinen  \nContents  \nAbstract 3  \nPreface 4  \nContents 5  \n1 Introduction 7  \n2 Background 7  \n2. 1 Postulates of Quantum Mechanics . . . . . . . . . . . . . . . . . . 8  \n2.2 Quantum Information Processing . . . . . . . . . . . . . . . . . . . 8  \n2.2. 1 Qubits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9  \n2.2.2 Quantum Circuits . . . . . . . . . . . . . . . . . . . . . . . 10  \n2.3 Quantum Computer Architectures .................. 12  \n2.4 Introduction to Machine Learning .................. 14  \n2.5 Classical Machine Learning ..................... 15  \n2.5. 1 Feedforward Neural Networks . . . . . . . . . . . . . . . . 16  \n2.5.2 Generative Adversarial Networks . . . . . . . . . . . . . . 17  \n2.5.3 Variational Autoencoders . . . . . . . . . . . . . . . . . . . 18  \n2.6 Quantum Machine Learning ..................... 19  \n2.6.1 Supervi","cbCaiiElb6gsOd6d","https://ap.wps.com/l/cbCaiiElb6gsOd6d","pdf",4239185,1,58,"English","en",105,"# Abstract\n# Preface\n# 1 Introduction\n# 2 Background\n## 2.1 Postulates of Quantum Mechanics\n## 2.2 Quantum Information Processing\n## 2.3 Quantum Computer Architectures\n## 2.4 Introduction to Machine Learning\n## 2.5 Classical Machine Learning\n## 2.6 Quantum Machine Learning\n## 2.7 Quantum vs Classical Machine Learning\n## 2.8 Molecule Representations\n## 2.9 Generating Molecules with Classical Machine Learning\n# 3 Methods-Generating Polymers with Quantum Machine Learning\n## 3.1 Representing Monomers and Polymers\n## 3.2 Design of a Baseline Classical Generator\n## 3.3 Estimating Properties of the Generated Polymers\n## 3.4 Design of a Quantum Generator\n## 3.5 Co-designing the Quantum Circuits\n# 4 Training Methods\n## 4.1 Property Estimator\n## 4.2 Classical Generators\n## 4.3 Quantum Generators Without Noise\n## 4.4 HPC for Quantum Generators\n## 4.5 Challenges of Parallelized Noisy Training","[{\"question\":\"What problem does the thesis address in polymer design?\",\"answer\":\"It targets inverse polymer design as a goal-oriented generation task, using machine learning to generate polymers that satisfy desired properties.\"},{\"question\":\"Which polymers and models are investigated?\",\"answer\":\"The study focuses on polyhydroxyalkanoates (PHAs) and compares quantum and classical generative models for producing polymers with targeted characteristics.\"},{\"question\":\"How is the impact of quantum noise evaluated?\",\"answer\":\"Quantum models are trained with depolarizing noise using different noise probabilities, and performance is assessed via test loss plus validity, novelty, and uniqueness benchmarks.\"}]","Inverse Polymer Design using Quantum Machine Learning - Master’s Thesis 2024 | PDF",1785723947,146,{"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},"inverse-polymer-design-using-quantum-machine-learning-masters-thesis-2024","",{"@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/inverse-polymer-design-using-quantum-machine-learning-masters-thesis-2024/119367/",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 thesis address in polymer design?","Question",{"text":75,"@type":76},"It targets inverse polymer design as a goal-oriented generation task, using machine learning to generate polymers that satisfy desired properties.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"Which polymers and models are investigated?",{"text":80,"@type":76},"The study focuses on polyhydroxyalkanoates (PHAs) and compares quantum and classical generative models for producing polymers with targeted characteristics.",{"name":82,"@type":73,"acceptedAnswer":83},"How is the impact of quantum noise evaluated?",{"text":84,"@type":76},"Quantum models are trained with depolarizing noise using different noise probabilities, and performance is assessed via test loss plus validity, novelty, and uniqueness benchmarks.","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"]