[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-127815-en":3,"doc-seo-127815-105":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},127815,1099523882367,"Hazel","https://ap-avatar.wpscdn.com/davatar_9964176cb1d06d4a9deccf72a44ae3dc",8,"Research & Report","Mixing predictions for online metric algorithms - arXiv 2304.01781v2 - Abstract and introduction","A major technique in learning-augmented online algorithms combines multiple predictors or algorithms. Because each predictor’s quality can change over time, the work seeks performance against dynamic combinations that follow different predictors at different moments, rather than only the single best predictor in hindsight. For metrical task systems, it achieves an O(ℓ2) competitive ratio versus an unconstrained combination of ℓ predictors, and shows this bound is tight. With limited predictor switches, it attains (1+ε)-competitiveness and also adapts to bandit-style access.","Mixing predictions for online metric algorithms  \narXiv :2304 .01781v2 [ cs .LG] 15 Dec 2023  \nAntonios Antoniadis University of Twente  \nChristian Coester University of Oxford  \nMarek Eli􀀓a􀀔s Bocconi University  \nAdam Polak  \nMax Planck Institute for Informatics  \nBertrand Simon IN2P3 Computing Center / CNRS  \nAbstract  \nA major technique in learning-augmented online algorithms is combining multiple algorithms or predictors. Since the performance of each predictor may vary over time, it is desirable to use not the single best predictor as a benchmark, but rather a dynamic combination which follows di􀀋erent predictors at di􀀋erent times. We design algorithms that combine predictions and are competitive against such dynamic combinations for a wide class of online problems, namely, metrical task systems. Against the best (in hindsight) unconstrained combination of ℓ predictors, we obtain a competitive ratio of O(ℓ2 ), and show that this is best possible. However, for a benchmark with slightly constrained number of switches between di􀀋erent predictors, we can get a (1 + ǫ)-competitive algorithm. Moreover, our algorithm can be adapted to access predictors in a bandit-like fashion, querying only one predictor at a time. An unexpected implication of one of our lower bounds is a new structural insight about covering formulations for the k-server problem.  \n1 Introduction  \nMotivated by the power of machine-learned predictions, the 􀀌eld of learningaugmented algorithms has been growing rapidly in recent years. In the classical 􀀌eld of online algorithms, an input sequence is revealed to an algorithm overtime and it is assumed that at all times, no information about the future part of the input is available. In contrast, a learning-augmented algorithm additionally has access to predictions (e.g., machine-learned) related to the future input. These predictions may be inaccurate, so a challenge is to simultaneously utilize high-quality predictions to their best advantage while at the same time avoiding to be misled by erroneous predictions.  \nAn important technique in the 􀀌eld of learning-augmented algorithms is the method of combining multiple algorithms into a single hybrid algorithm that leverages the advantages of all individual algorithms. The basic idea goes back to several decades before the area of learning-augmented algorithms was born and also has applications, for example, in pure online algorithms: Fiat et al. [1990] de􀀌ned a MIN operator on algorithms for the k-server problem that combines several algorithms into one whose cost matches the best of them up to  \na constant factor, and they used this technique to obtain the 􀀌rst competitive algorithm for the k-server problem.  \nIn learning-augmented algorithms, similar combination techniques are employed for several purposes. Firstly, they are frequently used to make algorithms robust against prediction errors by combining an algorithm that mostly follows predictions with a classical online algorithm that ignores predictions (see, e.g. , Lykouris and Vassilvitskii [2021], Purohit et al. [2018], Rohatgi [2020], Antoniadis et al. [2020], Wei [2020], Bamas et al. [2020], Bansal et al. [2022]) . Infact, one might argue that almost all algorithms that utilize predictions while being robust to their error are at least implicitly a kind of combination of two algorithms. A second purpose, as employed in Antoniadis et al. [2021], is to combine several di􀀋erently parameterized versions of the same algorithm in order to perform nearly as well as the version with the best parameter choice. These aforementioned works, like the majority of research in learning-augmented algorithms, focus on settings where a single predictor provides suggestions to the algorithm.  \nHowever, a third and perhaps the most relevant application of combining several algorithms in the learning-augmented realm is to be able to deal with multiple predictors. In practice it is often the case that several predictors are av","cbCaicdDZZULasy5","https://ap.wps.com/l/cbCaicdDZZULasy5","pdf",320310,1,24,"English","en",105,"# Introduction\n## Learning-augmented online algorithms and prediction errors\n## Combining algorithms for robustness and parameter tuning\n## Multiple predictors and dynamic combinations\n## Motivation: covering and k-server structural insights","[{\"question\":\"What problem setting motivates mixing predictions in online metric algorithms?\",\"answer\":\"The paper targets learning-augmented online algorithms where predictions about future inputs may be inaccurate and multiple predictors may conflict over time.\"},{\"question\":\"What competitive guarantee is obtained for metrical task systems versus an unconstrained combination?\",\"answer\":\"Against the best in-hindsight unconstrained combination of ℓ predictors, the paper proves a competitive ratio of O(ℓ2) and states it is best possible.\"},{\"question\":\"How does the approach improve when the number of predictor switches is constrained?\",\"answer\":\"With a benchmark that restricts switches between predictors, the algorithm achieves (1+ε)-competitiveness instead of the O(ℓ2) bound.\"}]","Mixing predictions for online metric algorithms - arXiv 2304.01781v2 - Abstract and introduction | PDF",1785942015,60,{"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},"mixing-predictions-for-online-metric-algorithms-arxiv-230401781v2-abstract-and-introduction","",{"@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/mixing-predictions-for-online-metric-algorithms-arxiv-230401781v2-abstract-and-introduction/127815/",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","2026-08-24","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},"What problem setting motivates mixing predictions in online metric algorithms?","Question",{"text":76,"@type":77},"The paper targets learning-augmented online algorithms where predictions about future inputs may be inaccurate and multiple predictors may conflict over time.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"What competitive guarantee is obtained for metrical task systems versus an unconstrained combination?",{"text":81,"@type":77},"Against the best in-hindsight unconstrained combination of ℓ predictors, the paper proves a competitive ratio of O(ℓ2) and states it is best possible.",{"name":83,"@type":74,"acceptedAnswer":84},"How does the approach improve when the number of predictor switches is constrained?",{"text":85,"@type":77},"With a benchmark that restricts switches between predictors, the algorithm achieves (1+ε)-competitiveness instead of the O(ℓ2) bound.","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},[94,98,102,106,110,115,120,123,128,131,135],{"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":29,"slug":109},5,"Comic","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":107,"slug":138},19,"General","general"]