[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-120455-en":3,"doc-seo-120455-105":29,"detail-sidebar-cat-0-en-105":90},{"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":11,"language":21,"language_code":22,"site_id":23,"html_lang":22,"table_of_contents":24,"faqs":25,"seo_title":26,"seo_description":14,"update_tm":27,"read_time":28},120455,962075114765,"Quinn","https://ap-avatar.wpscdn.com/davatar_a8503ba1806abce46bf441b54a3ca4cd",8,"Research & Report","A Comprehensive Sustainable Framework for Machine Learning and Artificial Intelligence","In financial contexts, regulations and best practices impose machine-learning requirements across four pillars: fairness, privacy, interpretability, and greenhouse-gas emissions. Although prior research addresses these topics individually, it rarely evaluates them jointly, despite fundamental trade-offs such as accuracy vs fairness or accuracy vs privacy. The paper introduces a Sustainable Machine Learning framework and the FPIG general AI pipeline, enabling simultaneous consideration of all pillars while learning their interactions. It further presents a meta-learning algorithm to estimate pillars from dataset summaries, architecture, and hyperparameters before training, supporting architecture selection under user requirements and demonstrating trade-offs on multiple datasets.","A Comprehensive Sustainable Framework for Machine Learning and Artificial Intelligence  \nRoberto Pagliaria, Peter Hilla, Po-Yu Chen a,b, * , Maciej Dabrownya, Tingsheng Tana and  \nFrancois Buet-Golfousea  \naJPMorgan  \nb Imperial College London  \narXiv :2407 . 12445v 1 [ cs .LG] 17 Jul 2024  \nAbstract. In financial applications, regulations or best practices often lead to specific requirements in machine learning relating to four key pillars: fairness, privacy, interpretability and greenhouse gas emissions. These all sit in the broader context of sustainability in AI, an emerging practical AI topic. However, although these pillars have been individually addressed by past literature, none of these works have considered all the pillars. There are inherent trade-offs between each of the pillars (for example, accuracy vs fairness or accuracy vs privacy), making it even more important to consider them together. This paper outlines a new framework for Sustainable Machine Learning and proposes FPIG, a general AI pipeline that allows for these critical topics to be considered simultaneously to learn the trade-offs between the pillars better. Based on the FPIG framework, we propose a meta-learning algorithm to estimate the four key pillars given a dataset summary, model architecture, and hyperparameters before model training. This algorithm allows users to select the optimal model architecture for a given dataset and a given set of user requirements on the pillars. We illustrate the trade-offs under the FPIG model on three classical datasets and demonstrate the meta-learning approach with an example of real-world datasets and models with different interpretability, showcasing how it can aid model selection.  \n1 Introduction  \nArtificial Intelligence has become an emerging tool essential for all financial sectors [37, 61, 53, 57] .  \nHowever, the characterisation of AI extends beyond the realm of technology and permeates into the precincts of infrastructure [26] and ideology [44], leading to an opacity around the concept of AI [40] . This nebulous nature of AI magnifies the challenges of effectively understanding and governing it while underscoring the need for malleability and interdisciplinary dialogue in AI ethics and governance. Consequently, this discourse does not gravitate towards a rigid definition of AI; rather, it embraces its polysemous essence and explores AI as a complex system [16] .  \nThe current landscape of AI ethics frameworks [32, 39] is peppered with a proliferation of proposed principles and a conspicuous absence of uniformity across these frameworks. The initial environmental rights and climate justice movements were driven by the United Nations Climate Change Conferences, and Sustainable Development Goals (SDGs) [6], along with the environmental, social and corporate governance (ESG) frameworks [4]. Unfortunately,  \n∗ [Corresponding Author. Email: po-yu.chen11@imperial.ac.uk](Corresponding Author. Email: po-yu.chen11@imperial.ac.uk).  \nwhile the environmental implications of AI are gradually entering the discourse [56], the broader concept of sustainability in AI appears to be largely overlooked [34] . Recent literature only fostered a narrow vision of sustainable AI [58, 62], neglecting the interconnected nature of various AI governance challenges. A holistic view of sustainable AI should be an amalgamation of three intertwined pillars: economic, environmental and social, necessitating a complex systems approach [27] .  \nFinancial institutions have particular duties in relation to AI that need to be paid close attention to. The Information Commissioner’s Office (ICO) has strict guidance on AI regarding interpretability, data protection and privacy [8] . Additionally, there have been several recent developments from significant organisations relating to AI regulations, bolstering the importance of sustainable AI’s key features. For example, the European Union has proposed the AI Act [7], a European law on AI. The B","cbCaispRHLm9Zrm2","https://ap.wps.com/l/cbCaispRHLm9Zrm2","pdf",1162354,1,"English","en",105,"# Introduction\n## AI ethics frameworks and sustainability challenges\n## Financial regulations and governance requirements\n## Sustainable Machine Learning and FPIG framework\n## Meta-learning for pillar estimation and model selection","[{\"question\":\"Which four pillars does the proposed sustainable AI framework focus on?\",\"answer\":\"The framework centers on fairness, privacy, interpretability, and greenhouse gas (GHG) emissions in machine-learning systems.\"},{\"question\":\"Why is a single unified framework important for these pillars?\",\"answer\":\"Because the pillars involve inherent trade-offs, addressing them separately can miss how changes in one objective affect the others, so they must be considered together.\"},{\"question\":\"How does the FPIG-based approach support model selection before training?\",\"answer\":\"It uses a meta-learning algorithm that estimates the four pillar outcomes from dataset summary, model architecture, and hyperparameters, allowing users to choose an architecture that best matches their requirements.\"}]","A Comprehensive Sustainable Framework for Machine Learning and Artificial Intelligence | PDF",1785730191,20,{"code":4,"msg":30,"data":31},"ok",{"site_id":23,"language":22,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":85,"head_meta":87,"extra_data":89,"updated_unix":27},"a-comprehensive-sustainable-framework-for-machine-learning-and-artificial-intelligence","",{"@graph":35,"@context":84},[36,53,67],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,50],{"item":40,"name":41,"@type":42,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":44,"name":45,"@type":42,"position":46},"https://docshare.wps.com/document/","Document",2,{"item":48,"name":12,"@type":42,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":42,"position":52},"https://docshare.wps.com/document/a-comprehensive-sustainable-framework-for-machine-learning-and-artificial-intelligence/120455/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":22,"description":14,"dateModified":61,"datePublished":61,"encodingFormat":60,"isAccessibleForFree":62,"interactionStatistic":63},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":40,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-08-03",true,{"@type":64,"interactionType":65,"userInteractionCount":4},"InteractionCounter",{"@type":66},"ViewAction",{"@type":68,"mainEntity":69},"FAQPage",[70,76,80],{"name":71,"@type":72,"acceptedAnswer":73},"Which four pillars does the proposed sustainable AI framework focus on?","Question",{"text":74,"@type":75},"The framework centers on fairness, privacy, interpretability, and greenhouse gas (GHG) emissions in machine-learning systems.","Answer",{"name":77,"@type":72,"acceptedAnswer":78},"Why is a single unified framework important for these pillars?",{"text":79,"@type":75},"Because the pillars involve inherent trade-offs, addressing them separately can miss how changes in one objective affect the others, so they must be considered together.",{"name":81,"@type":72,"acceptedAnswer":82},"How does the FPIG-based approach support model selection before training?",{"text":83,"@type":75},"It uses a meta-learning algorithm that estimates the four pillar outcomes from dataset summary, model architecture, and hyperparameters, allowing users to choose an architecture that best matches their requirements.","https://schema.org",{"og:url":51,"og:type":86,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":88,"canonical":51},"index,follow",{"doc_id":7,"site_id":23},{"code":4,"msg":5,"data":91},[92,96,100,104,109,114,119,122,126,129,133],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":93,"show_sort_weight":94,"slug":95},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":97,"show_sort_weight":98,"slug":99},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":45,"category_name":101,"show_sort_weight":102,"slug":103},"Exam",70,"exam",{"id":105,"doc_module":4,"doc_module_name":45,"category_name":106,"show_sort_weight":107,"slug":108},5,"Comic",60,"comic",{"id":110,"doc_module":4,"doc_module_name":45,"category_name":111,"show_sort_weight":112,"slug":113},6,"Technology",50,"technology",{"id":115,"doc_module":4,"doc_module_name":45,"category_name":116,"show_sort_weight":117,"slug":118},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":45,"category_name":124,"show_sort_weight":28,"slug":125},9,"Religion & Spirituality","religion-spirituality",{"id":28,"doc_module":4,"doc_module_name":45,"category_name":127,"show_sort_weight":28,"slug":128},"World Cup","world-cup",{"id":130,"doc_module":4,"doc_module_name":45,"category_name":131,"show_sort_weight":130,"slug":132},10,"Lifestyle","lifestyle",{"id":134,"doc_module":4,"doc_module_name":45,"category_name":135,"show_sort_weight":105,"slug":136},19,"General","general"]