[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-125403-en":3,"doc-seo-125403-105":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},125403,1374391974564,"Clementine","https://ap-avatar.wpscdn.com/avatar/14000253aa45c000a9e?x-image-process=image/resize,m_fixed,w_180,h_180&k=1779874745381141002",8,"Research & Report","Evaluating the Energy Consumption of Machine Learning - Systematic Literature Review and Experiments","Monitoring, understanding, and optimizing the energy consumption of Machine Learning (ML) are key motivations for evaluating energy use, yet no universal tool exists across all use cases. Evaluation can also differ by method and assumptions, making tool selection difficult. This work addresses the gap with a systematic literature review covering training and inference energy evaluation tools and methods, and an experimental protocol that compares selected approaches. The comparison combines qualitative and quantitative results across vision and language ML tasks, and the study provides open-source repositories for extending the review and augmenting experiments.","arXiv :2408 . 15128v1 [ cs .LG] 27 Aug 2024  \nEvaluating the Energy Consumption of Machine Learning: Systematic Literature Review and Experiments  \nCharlotte Rodriguez 1, 2 , Laura Degioanni 1 , Laetitia Kameni 1 , Richard Vidal 1 , and Giovanni  \nNeglia2  \n1 Accenture Labs, Sophia Antipolis, [France. Email: {firstname.lastname}@accenture.com](France. Email: {firstname.lastname}@accenture.com)[ ](France. Email: {firstname.lastname}@accenture.com)2 Inria Université Côte d’Azur, Sophia Antipolis, [France. Email: {firstname.lastname}@inria.fr](France. Email: {firstname.lastname}@inria.fr)  \nAugust 28, 2024  \nAbstract  \nMonitoring, understanding, and optimizing the energy consumption of Machine Learning (ML) are various reasons why it is necessary to evaluate the energy usage of ML. However, there exists no universal tool that can answer this question for all use cases, and there may even be disagreement on how to evaluate energy consumption for a specific use case. Tools and methods are based on different approaches, each with their own advantages and drawbacks, and they need to be mapped out and explained in order to select the most suitable one for a given situation. We address this challenge through two approaches. First, we conduct a systematic literature review of all tools and methods that permit to evaluate the energy consumption of ML (both at training and at inference), irrespective of whether they were originally designed for machine learning or general software. Second, we develop and use an experimental protocol to compare a selection of these tools and methods. The comparison is both qualitative and quantitative on a range of ML tasks of different nature (vision, language) and computational complexity. The systematic literature review serves as a comprehensive guide for understanding the array of tools and methods used in evaluating energy consumption of ML, for various use cases going from basic energy monitoring to consumption optimization. Two open-source repositories are provided for further exploration. The first one contains tools that can be used to replicate this work or extend the current review. The second repository houses the experimental protocol, allowing users to augment the protocol with new ML computing tasks and additional energy evaluation tools.  \nContents  \n1 Introduction 3  \n1.1 Background ........................................... 3  \n1.2 Research Question and Contributions ............................. 4  \n1.3 Paper Overview and Outline .................................. 5  \n2 Protocol of the Review 6  \n2.1 Collection of a Pool of Items .................................. 6  \n2.2 Selection of the Items ...................................... 7  \n2.3 Classification of the Selected Items and Data Extraction .................... 8  \n3 Execution of the Protocol 10  \n4 Overview and Summary of the Selected Items 12  \n4.1 Taxonomy ............................................ 12  \n4.2 Summary of the Selected Methods and Tools ......................... 16  \n4.3 Summary of Selected Surveys ................................. 27  \n5 Experimental Comparison of a Subset of Methods and Tools 31  \n5.1 ML Computing Tasks ...................................... 31  \n5.2 Experiments ........................................... 32  \n5.3 Observations .......................................... 34  \n6 Conclusion and Outlook 35  \nAcronyms 38  \nReferences 39  \nAppendices 49  \nA Appendix 49  \nA.1 Additional Tools not Documented within a Scientific Article ................. 49  \nA.2 Search Queries ......................................... 49  \nA.3 List of excluding words ..................................... 50  \nA.4 Full URLs ............................................ 51  \n1 Introduction  \nReducing the energy consumption of Machine Learning (ML) and Software in general has many motivations. Besides the environmental impact of computing, other factors include the actual cost of energy [24], and the energy limitations of b","cbCait2zfTs7w3Sx","https://ap.wps.com/l/cbCait2zfTs7w3Sx","pdf",699882,1,52,"English","en",105,"# Introduction\n## Background\n## Research Question and Contributions\n## Paper Overview and Outline\n# Protocol of the Review\n## Collection of a Pool of Items\n## Selection of the Items\n## Classification of the Selected Items and Data Extraction\n# Execution of the Protocol\n# Overview and Summary of the Selected Items\n## Taxonomy\n## Summary of the Selected Methods and Tools\n## Summary of Selected Surveys\n# Experimental Comparison of a Subset of Methods and Tools\n## ML Computing Tasks\n## Experiments\n## Observations\n# Conclusion and Outlook","[{\"question\":\"Why is evaluating ML energy consumption challenging across use cases?\",\"answer\":\"There is no universal tool that works for all scenarios, and evaluation practices may vary in how energy use is measured for a given use case.\"},{\"question\":\"What does the systematic literature review cover in this work?\",\"answer\":\"It surveys tools and methods that evaluate ML energy consumption for both training and inference, regardless of whether they were originally built for ML or general software.\"},{\"question\":\"How are the tools and methods compared in the experiments?\",\"answer\":\"An experimental protocol performs qualitative and quantitative comparisons across different ML tasks, including vision and language workloads with varying computational complexity.\"}]","Evaluating the Energy Consumption of Machine Learning - Systematic Literature Review and Experiments | PDF",1785898697,131,{"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},"evaluating-the-energy-consumption-of-machine-learning-systematic-literature-review-and-experiments","",{"@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/evaluating-the-energy-consumption-of-machine-learning-systematic-literature-review-and-experiments/125403/",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},"Why is evaluating ML energy consumption challenging across use cases?","Question",{"text":75,"@type":76},"There is no universal tool that works for all scenarios, and evaluation practices may vary in how energy use is measured for a given use case.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What does the systematic literature review cover in this work?",{"text":80,"@type":76},"It surveys tools and methods that evaluate ML energy consumption for both training and inference, regardless of whether they were originally built for ML or general software.",{"name":82,"@type":73,"acceptedAnswer":83},"How are the tools and methods compared in the experiments?",{"text":84,"@type":76},"An experimental protocol performs qualitative and quantitative comparisons across different ML tasks, including vision and language workloads with varying computational complexity.","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"]