[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-127517-en":3,"doc-seo-127517-105":30,"detail-sidebar-cat-0-en-105":92},{"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},127517,13056712833777,"Logic","https://ap-avatar.wpscdn.com/davatar_29158cc5080c5b710cf443261637dec0",8,"Research & Report","Towards Understanding Fairness and its Composition in Ensemble Machine Learning - Paper","Machine learning software is widely deployed, yet it can introduce fairness harms affecting minority groups by race, sex, age, and other attributes. While many methods measure and mitigate bias for single classifiers, real systems often rely on ensembles whose fairness emerges through the interaction and combination of multiple learners. This study investigates how fairness composes in popular ensemble algorithms—bagging, boosting, stacking, and voting—using 168 Kaggle-based models across four fairness datasets. Results show ensembles can be made fairer without mitigation techniques, and reveal how ensemble hyperparameters and data characteristics jointly shape the final fairness.","Towards Understanding Fairness and its Composition in Ensemble Machine Learning  \nUsman Gohar  \nDept. of Computer Science Iowa State University Ames, IA, USA [ugohar@iastate.edu](ugohar@iastate.edu)  \nSumon Biswas  \nSchool of Computer Science Carnegie Mellon University Pittsburgh, PA, USA [sumonb@cs.cmu.edu](sumonb@cs.cmu.edu)  \nHridesh Rajan  \nDept. of Computer Science Iowa State University Ames, IA, USA [hridesh@iastate.edu](hridesh@iastate.edu)  \narXiv :2212 .04593v2 [ cs .LG] 25 Apr 2023  \nAbstract— Machine Learning (ML) software has been widely adopted in modern society, with reported fairness implications for minority groups based on race, sex, age, etc. Many recent works have proposed methods to measure and mitigate algorithmic bias in ML models. The existing approaches focus on single classifierbased ML models. However, real-world ML models are often composed of multiple independent or dependent learners in an ensemble (e.g., Random Forest), where the fairness composes in anon-trivial way. How does fairness compose in ensembles? What are the fairness impacts of the learners on the ultimate fairness of the ensemble? Can fair learners result in an unfair ensemble? Furthermore, studies have shown that hyperparameters influence the fairness of ML models. Ensemble hyperparameters are more complex since they affect how learners are combined in different categories of ensembles. Understanding the impact of ensemble hyperparameters on fairness will help programmers design fair ensembles. Today, we do not understand these fully for different ensemble algorithms. In this paper, we comprehensively study popular real-world ensembles: Bagging, Boosting, Stacking, and Voting. We have developed a benchmark of 168 ensemble models collected from Kaggle on four popular fairness datasets. We use existing fairness metrics to understand the composition of fairness. Our results show that ensembles can be designed to be fairer without using mitigation techniques. We also identify the interplay between fairness composition and data characteristics to guide fair ensemble design. Finally, our benchmark can be leveraged for further research on fair ensembles. To the best of our knowledge, this is one of the first and largest studies on fairness composition in ensembles yet presented in the literature. Index Terms—fairness, ensemble, machine learning, models  \nI. INTRODUCTION  \nMachine learning (ML) is ubiquitous in modern software today. Due to the black-box [1] nature of ML algorithms and its applications in critical decision-making [2, 3], fairness in ML software has become a huge concern. Measuring ML fairness [4–7] and mitigating the discrimination [5, 8, 9] has been studied extensively. Recent work in software engineering has shown the need to produce fair software and detect bias in complex ML software environments [10–13] .  \nPrior research has mostly focused on fairness in standalone classifiers (e.g., Logistic Regression, SVM) [1, 14, 15] . However, a class of ML models called ensemble models are becoming increasingly important in practice today due to their superior performance across a multitude of ML & real-life challenges [16–20], and better generalization on unseen data,  \nespecially in smaller datasets [18, 21, 22] . Ensemble models combine the predictions of multiple base learners to make the final prediction, e.g., Random Forest uses a large number of decision trees, with the majority class being the final output. Ensemble models are the most mentioned ML algorithms on Kaggle [23], and in previous SE works on fairness, ensemble models comprise more than 80% of the total models [12, 13] . Like traditional ML models, ensemble models can also suffer from unfairness problem that discriminates against population subgroups based on race, gender, etc. Although many fairness mitigation techniques [24, 25] exist, they do not always generalize well [26–28] . Therefore, if we better understand the fairness composition in ensembles, we can design","cbCair1zSjX5jbvp","https://ap.wps.com/l/cbCair1zSjX5jbvp","pdf",1539649,1,13,"English","en",105,"# Introduction\n## Fairness in Machine Learning\n## Limitations of Standalone Classifiers\n## Role of Ensemble Models in Fairness\n## Related Work on Fairness in Ensembles\n# Benchmark and Study Overview\n## Ensemble Algorithms Studied\n## Fairness Datasets and Model Collection\n## Fairness Metrics and Evaluation Approach","[{\"question\":\"Why is fairness harder to analyze for ensemble machine learning than for single classifiers?\",\"answer\":\"Ensemble predictions combine multiple base learners, so the resulting fairness depends on how learners interact rather than on one classifier in isolation.\"},{\"question\":\"Which ensemble algorithms are studied in the paper?\",\"answer\":\"The paper comprehensively studies bagging, boosting, stacking, and voting.\"},{\"question\":\"What is the benchmark used to evaluate fairness composition in ensembles?\",\"answer\":\"The study uses a benchmark of 168 ensemble models collected from Kaggle across four popular fairness datasets.\"}]","Towards Understanding Fairness and its Composition in Ensemble Machine Learning - Paper | PDF",1785939688,33,{"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":87,"head_meta":89,"extra_data":91,"updated_unix":28},"towards-understanding-fairness-and-its-composition-in-ensemble-machine-learning-paper","",{"@graph":36,"@context":86},[37,54,69],{"@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/towards-understanding-fairness-and-its-composition-in-ensemble-machine-learning-paper/127517/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-22","2026-08-05",true,{"@type":66,"interactionType":67,"userInteractionCount":20},"InteractionCounter",{"@type":68},"ViewAction",{"@type":70,"mainEntity":71},"FAQPage",[72,78,82],{"name":73,"@type":74,"acceptedAnswer":75},"Why is fairness harder to analyze for ensemble machine learning than for single classifiers?","Question",{"text":76,"@type":77},"Ensemble predictions combine multiple base learners, so the resulting fairness depends on how learners interact rather than on one classifier in isolation.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"Which ensemble algorithms are studied in the paper?",{"text":81,"@type":77},"The paper comprehensively studies bagging, boosting, stacking, and voting.",{"name":83,"@type":74,"acceptedAnswer":84},"What is the benchmark used to evaluate fairness composition in ensembles?",{"text":85,"@type":77},"The study uses a benchmark of 168 ensemble models collected from Kaggle across four popular fairness datasets.","https://schema.org",{"og:url":52,"og:type":88,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":90,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":93},[94,98,102,106,111,116,121,124,129,132,136],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":95,"show_sort_weight":96,"slug":97},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":99,"show_sort_weight":100,"slug":101},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":103,"show_sort_weight":104,"slug":105},"Exam",70,"exam",{"id":107,"doc_module":4,"doc_module_name":46,"category_name":108,"show_sort_weight":109,"slug":110},5,"Comic",60,"comic",{"id":112,"doc_module":4,"doc_module_name":46,"category_name":113,"show_sort_weight":114,"slug":115},6,"Technology",50,"technology",{"id":117,"doc_module":4,"doc_module_name":46,"category_name":118,"show_sort_weight":119,"slug":120},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":122,"slug":123},30,"research-report",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":126,"show_sort_weight":127,"slug":128},9,"Religion & Spirituality",20,"religion-spirituality",{"id":127,"doc_module":4,"doc_module_name":46,"category_name":130,"show_sort_weight":127,"slug":131},"World Cup","world-cup",{"id":133,"doc_module":4,"doc_module_name":46,"category_name":134,"show_sort_weight":133,"slug":135},10,"Lifestyle","lifestyle",{"id":137,"doc_module":4,"doc_module_name":46,"category_name":138,"show_sort_weight":107,"slug":139},19,"General","general"]