[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-124557-en":3,"doc-seo-124557-105":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},124557,1099513958607,"Jiven","https://ap-avatar.wpscdn.com/avatar/100002390cf8733938c?x-image-process=image/resize,m_fixed,w_180,h_180&k=1778829742770036399",8,"Research & Report","Automated Machine Learning for Deep Recommender Systems - A Survey - Comprehensive Summary","Deep recommender systems (DRS) address information overload by recommending items matched to users’ interests through powerful feature representations and modeling of nonlinear user–item relationships. Despite their advances, DRS architectures rely on complex neural components typically designed and tuned manually by experts. This survey comprehensively summarizes automated machine learning (AutoML) methods for building DRS, covering system overview, state-of-the-art approaches, and automated feature selection, embeddings, interactions, and system design, then highlights future research directions and provides conclusions.","Automated Machine Learning for Deep Recommender Systems: A Survey  \nBo Chen 1 􀀃 , Xiangyu Zhao2 􀀃 , Yejing Wang2 , Wenqi Fan3 , Huifeng Guo 1 , Ruiming Tang 1  \n1Huawei Noah's Ark Lab, 2 City University of Hong Kong, 3The Hong Kong Polytechnic University  \nfchenbo116, huifeng.guo, [tangruiming](tangruimingg@huawei.com)[g](tangruimingg@huawei.com)[@huawei.com](tangruimingg@huawei.com), [xianzhao@cityu.edu.hk](xianzhao@cityu.edu.hk), [adave631@gmail.com](adave631@gmail.com),  \n[wenqifan03@gmail.com](wenqifan03@gmail.com)  \narXiv :2204 .0 1390v 1 [ cs .IR] 4 Apr 2022  \nAbstract  \nDeep recommender systems (DRS) are critical for current commercial online service providers, which address the issue of information overload by recommending items that are tailored to the user's interests and preferences. They have unprecedented feature representations effectiveness and the capacity of modeling the non-linear relationships between users and items. Despite their advancements, DRS models, like other deep learning models, employ sophisticated neural network architectures and other vital components that are typically designed and tuned by human experts. This article will give a comprehensive summary of automated machine learning (AutoML) for developing DRS models.  \nWe first provide an overview of AutoML for DRS models and the related techniques. Then we discuss the state-of-the-art AutoML approaches that automate the feature selection, feature embeddings, feature interactions, and system design in DRS.  \nFinally, we discuss appealing research directionsand summarize the survey.  \n1 Introduction  \nRecent years have witnessed the explosive growth of online service providers [Ricci et al., 2011], including a range of scenarios like movies, music, news, short videos, ecommerces, etc. This leads to the increasingly serious information overload issue, overwhelming web users. Recommender systems are effective mechanisms that mitigate the above issue by intelligently retrieving and suggesting personalized items, e.g., contents and products, to users in their information-seeking endeavors, so as to match their interests and requirements better. With the development and prevalence of deep learning, deep recommender systems (DRS) have piqued interests from both academia and industrial communities [Zhang et al., 2019; Nguyen et al., 2017], due to their superior capacity of learning feature representations and modeling non-linear interactions between users and items [Zhang et al., 2019] .  \n􀀃 Both authors contributed equally to this research.  \nTo construct DRS architectures, the most common practice is to design and tune the different components in a hand-crafted fashion. However, manual development is fraught with three inherent challenges. First, this requires extensive expertise in deep learning and recommender systems. Second, substantial engineering labor and time cost are required to design task-specific components for various recommendation scenarios. Third, human bias and error can result in suboptimal DRS components, further reducing recommendation performance.  \nRecently, powered by the advances of both theories and technologies in automated machine learning (AutoML), tremendous interests are emerging for automating the components of DRS. By involving AutoML for the deep recommender systems, different models can be automatically designed according to various data, thus improving the prediction performance and enhancing generalization. Besides, it is helpful to eliminate the negative influence for DRS from human bias and error, as well as reduce artificialand temporal costs significantly. As shown in Tabel 1, we summarize these researches from the perspective of search space and search strategy, which are two critical factors for AutoML. Typically, these works can be divided into the following categories according to the components in DRS:  \n• Feature Selection: This is the process of selecting a subset of the most predictive and relevant features ","cbCaiiw7cM23E5dg","https://ap.wps.com/l/cbCaiiw7cM23E5dg","pdf",436263,1,"English","en",105,"# Abstract\n# Introduction\n## Information overload and recommender systems\n## Why manual DRS design is challenging\n## AutoML for deep recommender systems\n## AutoML categories in DRS\n## Table 1: Automated Machine Learning for Deep Recommender Systems","[{\"question\":\"What problem do deep recommender systems (DRS) solve?\",\"answer\":\"DRS mitigate information overload by recommending items tailored to users’ interests and preferences.\"},{\"question\":\"Why is manual design of DRS components challenging?\",\"answer\":\"Manual development requires deep expertise, large engineering effort and time, and can introduce human bias and error that reduce recommendation performance.\"},{\"question\":\"Which aspects of DRS does the survey focus on for AutoML automation?\",\"answer\":\"It covers AutoML approaches that automate feature selection, feature embeddings, feature interactions, and system design for DRS models.\"}]","Automated Machine Learning for Deep Recommender Systems - A Survey - Comprehensive Summary | PDF",1785892985,20,{"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},"automated-machine-learning-for-deep-recommender-systems-a-survey-comprehensive-summary","",{"@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/automated-machine-learning-for-deep-recommender-systems-a-survey-comprehensive-summary/124557/",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","2026-08-05",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},"What problem do deep recommender systems (DRS) solve?","Question",{"text":74,"@type":75},"DRS mitigate information overload by recommending items tailored to users’ interests and preferences.","Answer",{"name":77,"@type":72,"acceptedAnswer":78},"Why is manual design of DRS components challenging?",{"text":79,"@type":75},"Manual development requires deep expertise, large engineering effort and time, and can introduce human bias and error that reduce recommendation performance.",{"name":81,"@type":72,"acceptedAnswer":82},"Which aspects of DRS does the survey focus on for AutoML automation?",{"text":83,"@type":75},"It covers AutoML approaches that automate feature selection, feature embeddings, feature interactions, and system design for DRS models.","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"]