[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-118118-en":3,"doc-seo-118118-105":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},118118,5909877438554,"Maeve","https://ap-avatar.wpscdn.com/avatar/5600025385ad2bf12a7?_k=1778553567797529272",8,"Research & Report","Causal Feature Selection for Responsible Machine Learning - A Survey","Machine learning models increasingly shape high-impact decisions, making responsible machine learning essential for aligning outputs with ethical and social values. This survey examines four pillars of responsible ML—interpretability, fairness, adversarial robustness, and domain generalization—and explains why feature selection is central to achieving them. It focuses on causal feature selection, contrasting causal impact with mere statistical correlation to avoid spurious, biased patterns that can undermine reliability and trustworthiness in complex, real-world settings.","Causal Feature Selection for Responsible Machine Learning  \nRaha Moraffah , Paras Sheth⋆ , Saketh Vishnubhatla∗ and Huan Liu  \nSchool of Computing and Augmented Intelligence, Arizona State University, USA {raha.moraffah, psheth5, svishnu6, [huanliu](huanliu}@asu.edu)[}](huanliu}@asu.edu)[@asu.edu](huanliu}@asu.edu)  \narXiv :2402 .02696v 1 [ cs .LG] 5 Feb 2024  \nAbstract  \nMachine Learning (ML) has become an integral aspect of many real-world applications. As a result, the need for responsible machine learning has emerged, focusing on aligning ML models to ethical and social values, while enhancing their reliability and trustworthiness. Responsible ML involves many issues. This survey addresses four main issues: interpretability, fairness, adversarial robustness, and domain generalization. Feature selection plays a pivotal role in the responsible ML tasks. However, building upon statistical correlations between variables can lead to spurious patterns with biases and compromised performance.  \nThis survey focuses on the current study of causal feature selection: what it is and how it can reinforce the four aspects of responsible ML. By identifying features with causal impacts on outcomes and distinguishing causality from correlation, causal feature selection is posited as a unique approach to ensuring ML models to be ethically and socially responsible in high-stakes applications.  \n1 Introduction  \nMachine learning (ML) has been deeply and quickly incorporated into many facets of contemporary life. These days, ML models are used for vital tasks like disaster relief [Berariu et al., 2015] and disinformation campaigns [Islam et al., 2020], in addition to improving routine tasks like multimedia content recommendations [Sheth et al., 2023] . But the rapid use of these technologies has also brought to light serious issues with their reliability, transparency, and ethical application [Xu and Saleh, 2021; Rengasamy et al., 2021] . The discipline of responsible machine learning was founded in response to these worries, with the goal of coordinating ML developments with social, legal, and ethical norms.  \nResponsible Machine Learning can be defined as an approach that focuses on creating models which not only achieve their intended task but do so in a way that is understandable and transparent, treats all individuals and groups equitably, remains robust against manipulative and deceptive  \n∗Equal Contribution  \ninputs, and performs reliably across a variety of different environments and contexts. This approach ensures that the technology not only serves its immediate purpose but does so in a manner that aligns with broader ethical principles and adapts effectively to diverse and changing conditions [Arrieta and others, 2020; Hall et al., 2020] . For instance, when ML systems are used for hiring processes they might have significant impacts in ensuring responsible decisions. Responsibility of decisions can be quantified across various aspects, including model outputs and model behavior in different situations. When considering model outputs, the responsible ML model should generate (1) interpretable and (2) fair decisions. Interpretability involves the ML system’s ability to provide clear, understandable explanations for its decisions [Carvalho and others, 2019] . For instance, if a candidate is not selected, the system should be able to articulate why, perhaps due to lacking specific skills relevant to the job. Fairness requires that ML models assess candidates equitably, without biases based on gender, ethnicity, or other unrelated characteristics, ensuring all applicants are evaluated solely on their qualifications [Mehrabi et al., 2021] .  \nResponsible ML models should also be robust to manipulations and generalizable in unseen scenarios. Adversarial Robustness is about the system’s resilience against manipulative inputs, e.g., when candidates overstate their qualifications [Silva and Najafirad, 2020] . This ensures the integrity and ","cbCaitFzxFaXxej9","https://ap.wps.com/l/cbCaitFzxFaXxej9","pdf",529787,1,9,"English","en",105,"# Introduction\n## Responsible Machine Learning: four pillars\n## Feature selection and its limitations\n## Causal feature selection and its role","[{\"question\":\"What is the main focus of this survey on responsible machine learning?\",\"answer\":\"It studies causal feature selection and how it can strengthen interpretability, fairness, adversarial robustness, and domain generalization in responsible ML.\"},{\"question\":\"Why can traditional feature selection be insufficient for responsible ML?\",\"answer\":\"It often relies on statistical correlations, which may produce spurious patterns and biases, reducing performance and weakening ethical reliability.\"},{\"question\":\"How does causal feature selection help distinguish causality from correlation?\",\"answer\":\"It identifies features with causal impacts on outcomes and separates genuine causal drivers from historically correlated signals that may reinforce bias or fail in new contexts.\"}]","Causal Feature Selection for Responsible Machine Learning - A Survey | PDF",1785681704,23,{"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},"causal-feature-selection-for-responsible-machine-learning-a-survey","",{"@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/causal-feature-selection-for-responsible-machine-learning-a-survey/118118/",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","2026-08-02",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 is the main focus of this survey on responsible machine learning?","Question",{"text":75,"@type":76},"It studies causal feature selection and how it can strengthen interpretability, fairness, adversarial robustness, and domain generalization in responsible ML.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"Why can traditional feature selection be insufficient for responsible ML?",{"text":80,"@type":76},"It often relies on statistical correlations, which may produce spurious patterns and biases, reducing performance and weakening ethical reliability.",{"name":82,"@type":73,"acceptedAnswer":83},"How does causal feature selection help distinguish causality from correlation?",{"text":84,"@type":76},"It identifies features with causal impacts on outcomes and separates genuine causal drivers from historically correlated signals that may reinforce bias or fail in new contexts.","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},[93,97,101,105,110,115,120,123,127,130,134],{"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":21,"doc_module":4,"doc_module_name":46,"category_name":124,"show_sort_weight":125,"slug":126},"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":46,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":46,"category_name":136,"show_sort_weight":106,"slug":137},19,"General","general"]