[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-125438-en":3,"doc-seo-125438-105":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},125438,687197207919,"Theodora","https://ap-avatar.wpscdn.com/avatar/a000253d6f5f7c60be?x-image-process=image/resize,m_fixed,w_180,h_180&k=1779446848396160552",8,"Research & Report","Combining machine learning and optimization algorithms in 5G networks - Master’s thesis","This project combines machine learning techniques with optimization heuristics to address an NP-hard network operation planning problem: finding an optimal network configuration for sets of dynamic demands. The work develops constructive heuristics, along with a local search and a path-relinking strategy tailored to this task. It further studies two scenarios where ML extracts information from heuristic solutions to improve performance, including dimensionality reduction using principal component analysis and autoencoders to learn useful features.","Master of Science in Advanced Mathematics and Mathematical Engineering  \nTitle: Combining machine learning and optimization algorithms in 5G networks  \nAuthor: Lluís Sabater Rojas  \nAdvisor: Marc Ruiz Ramírez  \nDepartment: Department of Mathematics  \nAcademic year: 2022-2023  \nUniversitat Polit`ecnica de Catalunya Facultat de Matem`atiques i Estad´ıstica  \nMaster in Advanced Mathematics and Mathematical Engineering  \nMaster’s thesis  \nCombining machine learning and optimization algorithms in 5G networks  \nLlu´ıs Sabater Rojas  \nSupervised by Marc Ruiz Ram´ırez  \nMay, 2023  \nI want to thank PhD Marc Ruiz Ram´ırez for his time and guidance through this project. His help has been essential to understand and produce the results of this project.  \nI also want to my family for their unconditional support.  \nAbstract  \nThis project aims to combine machine learning (ML) techniques and optimization heuristics for a complex operation in networks problem that requires finding the optimal network configuration for a set of dynamic demands. This problem is NP-hard, so we develop some constructive heuristics to solve it, we also describe a possible local search and path-relinking algorithm. These heuristics are handcrafted for this specific task and are the common way of tackling these type of problems. In this project, we also explore two scenarios where ML can be used to extract information from the heuristic solutions (via dimensionality reduction) and enhance the heuristic performance: one using multiple problem instances and another using the solutions of a single instance. We use principal component analysis (PCA) and autoencoders (AE) to reduce the solutions’ data dimensionality and try to learn useful features for the heuristics.  \nKeywords  \nAutoencoders, Combinatorial optimization, Heuristic algorithms, Integer linear programming, Machine learning, Networks, Principal component analysis.  \nContents  \n1 Introduction 3  \n1.1 Project structure ........................................ 4  \n2 Background 5  \n2.1 Optimization .......................................... 5  \n2.1.1 Integer Linear Programming ............................. 5  \n2.1.2 Constructive Heuristics and Local Search ....................... 6  \n2.2 Statistics and Machine Learning ............................... 7  \n2.2.1 Principal Component Analysis ............................ 7  \n2.2.2 Autoencoders ..................................... 9  \n2.3 Operation in Networks ..................................... 11  \n3 Problem formulation 14  \n3.1 Complexity of the problem .................................. 16  \n3.2 Heuristic approach ....................................... 17  \n3.2.1 Constructive heuristics ................................ 19  \n3.2.2 Local Search ...................................... 23  \n3.2.3 Path-relinking ..................................... 24  \n4 Combining ML for optimization 27  \n4.1 Use of Machine Learning in our problem ........................... 27  \n4.1.1 Use case 1: Multiple Instances ............................ 28  \n4.1.2 Use case 2: Single Instance .............................. 29  \n5 Results 30  \n5.1 Heuristic Results ........................................ 30  \n5.2 Results for use case 1 ..................................... 33  \n5.2.1 Using instances only .................................. 33  \n5.2.2 Using Best solutions .................................. 35  \n5.3 Results for use case 2 ..................................... 38  \n5.3.1 Other analysis ..................................... 42  \n6 Conclusions 46  \nA Appendix 51  \n1. Introduction  \nA telecommunication network requires continuous adjustment and adaptation to different needs and scenarios. The conventional approach to telecommunication network design was to predict the long-term future traffic and plan the network capacity accordingly. At the same time, a static network configuration was also planned in order to determine how the network responds to different situations. However,","cbCaivsFBEFH33Kk","https://ap.wps.com/l/cbCaivsFBEFH33Kk","pdf",3821539,1,65,"English","en",105,"# Introduction\n## Project structure\n# Background\n## Optimization\n## Statistics and Machine Learning\n## Operation in Networks\n# Problem formulation\n## Complexity of the problem\n## Heuristic approach\n# Combining ML for optimization\n## Use of Machine Learning in our problem\n# Results\n## Heuristic Results\n## Results for use case 1\n## Results for use case 2\n# Conclusions\n## Appendix","[{\"question\":\"What problem does the thesis focus on in 5G networks?\",\"answer\":\"It focuses on operation planning that requires selecting an optimal network configuration for dynamic demands while minimizing network capacity usage. The underlying optimization problem is NP-hard.\"},{\"question\":\"What heuristic methods are developed in the project?\",\"answer\":\"The thesis develops constructive heuristics and describes a local search and a path-relinking algorithm. These methods are handcrafted for the specific network task.\"},{\"question\":\"How is machine learning used to improve heuristic performance?\",\"answer\":\"Two scenarios are explored: using multiple problem instances or using solutions from a single instance. Dimensionality reduction with PCA and autoencoders is used to extract informative features from heuristic solutions.\"}]","Combining machine learning and optimization algorithms in 5G networks - Master’s thesis | PDF",1785898926,164,{"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},"combining-machine-learning-and-optimization-algorithms-in-5g-networks-masters-thesis","",{"@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/combining-machine-learning-and-optimization-algorithms-in-5g-networks-masters-thesis/125438/",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","2026-08-05",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 thesis focus on in 5G networks?","Question",{"text":75,"@type":76},"It focuses on operation planning that requires selecting an optimal network configuration for dynamic demands while minimizing network capacity usage. The underlying optimization problem is NP-hard.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What heuristic methods are developed in the project?",{"text":80,"@type":76},"The thesis develops constructive heuristics and describes a local search and a path-relinking algorithm. These methods are handcrafted for the specific network task.",{"name":82,"@type":73,"acceptedAnswer":83},"How is machine learning used to improve heuristic performance?",{"text":84,"@type":76},"Two scenarios are explored: using multiple problem instances or using solutions from a single instance. Dimensionality reduction with PCA and autoencoders is used to extract informative features from heuristic solutions.","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"]