[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-82621-en":3,"doc-seo-82621-105":28,"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":20,"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":13,"seo_description":14,"update_tm":26,"read_time":27},82621,8796095360427,"Lucas Martin","https://ap-avatar.wpscdn.com/davatar_994ba38a5ba835b3df7d355c54d3ed8d",8,"Research & Report","MMAO-Cls: Metabolic Multi-Agent Optimization for Joint Feature Selection and Classifier Tuning","MMAO-Cls studies whether the Metabolic Multi-Agent Optimizer (MMAO) can serve as a credible outer-loop optimizer for classification model selection. Each agent encodes a binary feature mask and classifier hyperparameters in a mixed space, while private energy, communal budget, role drift, and lifecycle turnover map to an accuracy–complexity tradeoff for wrapper learning. Feature-budget adaptation uses feature-information priors and validation reward regularizes subset compactness and train–validation overfitting gap. Experiments on seven tabular benchmarks show strong held-out transferability with compact subsets.","MMAO-Cls: Metabolic Multi-Agent Optimization for Joint Feature Selection and Classifier Tuning  \nJinliang Xu∗ and Liping Ma  \narXiv :2607 .0 1539v2 [ cs .NE] 4 Jul 2026  \nAbstract—This paper studies whether the Metabolic MultiAgent Optimizer (MMAO) can act as a credible outer-loop optimizer for classification model selection. We propose MMAOCls, a mixed-space realization in which each agent jointly encodes a binary feature mask and classifier hyperparameters, while private energy, communal budget, role drift, and lifecycle turnover are mapped to the accuracy-complexity tradeoff of wrapper learning. The implementation is strengthened by deriving feature-budget adaptation from feature-information priors and by regularizing validation reward with both subset compactness and train-validation overfitting gap. We evaluate MMAO-Cls on seven standard tabular benchmarks with three seeds each and compare it against RandomSearch, GA-lite, PSOlite, and an endogenous no-sharing ablation. On the aggregate validation objective, MMAO-Cls ranks second (0 .9433) behind GA-lite (0 .9446). On held-out test performance, it reaches meanscore 0.8882, improving over RandomSearch (0 .8808) and GAlite (0 .8857), remaining close to PSO-lite (0 .8874) and the nosharing ablation (0 .8900), while using the most compact mean held-out feature subset among all compared methods (feature ratio 0.4881). Pairwise tests show that these margins are not yet statistically significant. The strongest resulting claim is therefore transferability rather than outright superiority: MMAO-Cls supports classification applicability and compact mixed-space search more clearly than it isolates communal sharing as a decisive standalone advantage.  \nIndex Terms—Classification optimization, feature selection, hyperparameter tuning, wrapper methods, mixed-space metaheuristics, MMAO.  \nI. INTRODUCTION  \nCLASSIFICATION pipelines often rely on design deci  \nsions that are not learned directly by the base classifier. Feature subset selection, hyperparameter tuning, and complexity control usually remain outer-loop optimization problems. These problems are mixed by nature: a candidate pipeline may include binary feature-inclusion decisions, continuous or ordinal hyperparameters, and a nontrivial tradeoff between predictive quality and subset compactness [1]–[3] . This places the task close to mixed-variable black-box optimization and algorithm configuration, where recent studies emphasize both landscape heterogeneity and the practical difficulty of fair lowbudget comparison [4], [5] .  \nThis makes classification a useful testbed for the Metabolic Multi-Agent Optimizer (MMAO) . MMAO is not a classifier; it is a search-control framework built around a private-public metabolic economy. Agents earn or lose private energy, donate  \nJinliang Xu is an independent researcher in Beijing, China; e-mail: jlxu[fly@gmail.com](fly@gmail.com).  \nLiping Ma is with the Department of Disease Control and Prevention, The Seventh Medical Center of Chinese PLA General Hospital, Beijing, China; e-mail: [lipingmaqzx@163.com](lipingmaqzx@163.com).  \npart of successful gains to a communal pool, drift continuously between exploratory and exploitative roles, and are replaced when their search behavior stops paying for itself. The framework hypothesis is that heterogeneous search should emerge from this closed loop rather than from a stack of externally attached schedules.  \nThe classification setting is an informative stress test for that hypothesis. Compared with standard continuous benchmarks, model selection for classification is mixed-space, evaluationexpensive, and vulnerable to overfitting. If MMAO is genuinely reusable across domains, it should coordinate featuremask search and hyperparameter search under one explanatory controller while still producing sensible held-out performance.  \nThis paper develops that idea through MMAO-Cls. Each agent carries a mixed representation whose discrete part determines a fe","cbCainEP7tSUb15g","https://ap.wps.com/l/cbCainEP7tSUb15g","pdf",570272,1,"English","en",105,"# Abstract\n# Introduction\n# Related Work","[{\"question\":\"What problem does MMAO-Cls address in classification pipelines?\",\"answer\":\"It targets outer-loop decisions in classification, specifically joint feature subset selection and classifier hyperparameter tuning under an accuracy–complexity tradeoff.\"},{\"question\":\"How does MMAO-Cls represent and optimize candidates?\",\"answer\":\"Each agent uses a mixed representation: a discrete binary feature mask for subset selection and a continuous hyperparameter vector for classifier tuning, optimized via a metabolic multi-agent loop.\"},{\"question\":\"What do the benchmark results suggest about MMAO-Cls?\",\"answer\":\"On aggregate validation, it ranks near the top and on held-out test it improves over RandomSearch and GA-lite while remaining close to PSO-lite and a no-sharing ablation, supporting transferability rather than absolute superiority.\"}]",1784181860,20,{"code":4,"msg":29,"data":30},"ok",{"site_id":23,"language":22,"slug":31,"title":13,"keywords":32,"description":14,"schema_data":33,"social_meta":85,"head_meta":87,"extra_data":89,"updated_unix":26},"mmao-cls-metabolic-multi-agent-optimization-for-joint-feature-selection-and-classifier-tuning","",{"@graph":34,"@context":84},[35,52,67],{"@type":36,"itemListElement":37},"BreadcrumbList",[38,42,46,49],{"item":39,"name":40,"@type":41,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":43,"name":44,"@type":41,"position":45},"https://docshare.wps.com/document/","Document",2,{"item":47,"name":12,"@type":41,"position":48},"https://docshare.wps.com/document/research-report/",3,{"item":50,"name":13,"@type":41,"position":51},"https://docshare.wps.com/document/mmao-cls-metabolic-multi-agent-optimization-for-joint-feature-selection-and-classifier-tuning/82621/",4,{"url":50,"name":13,"@type":53,"author":54,"headline":13,"publisher":56,"fileFormat":59,"inLanguage":22,"description":14,"dateModified":60,"datePublished":61,"encodingFormat":59,"isAccessibleForFree":62,"interactionStatistic":63},"DigitalDocument",{"name":9,"@type":55},"Person",{"url":39,"name":57,"@type":58},"DocShare","Organization","application/pdf","2026-07-17","2026-07-16",true,{"@type":64,"interactionType":65,"userInteractionCount":20},"InteractionCounter",{"@type":66},"ViewAction",{"@type":68,"mainEntity":69},"FAQPage",[70,76,80],{"name":71,"@type":72,"acceptedAnswer":73},"What problem does MMAO-Cls address in classification pipelines?","Question",{"text":74,"@type":75},"It targets outer-loop decisions in classification, specifically joint feature subset selection and classifier hyperparameter tuning under an accuracy–complexity tradeoff.","Answer",{"name":77,"@type":72,"acceptedAnswer":78},"How does MMAO-Cls represent and optimize candidates?",{"text":79,"@type":75},"Each agent uses a mixed representation: a discrete binary feature mask for subset selection and a continuous hyperparameter vector for classifier tuning, optimized via a metabolic multi-agent loop.",{"name":81,"@type":72,"acceptedAnswer":82},"What do the benchmark results suggest about MMAO-Cls?",{"text":83,"@type":75},"On aggregate validation, it ranks near the top and on held-out test it improves over RandomSearch and GA-lite while remaining close to PSO-lite and a no-sharing ablation, supporting transferability rather than absolute superiority.","https://schema.org",{"og:url":50,"og:type":86,"og:title":13,"og:site_name":57,"og:description":14},"article",{"robots":88,"canonical":50},"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":44,"category_name":93,"show_sort_weight":94,"slug":95},"Story & Novel",90,"story-novel",{"id":45,"doc_module":4,"doc_module_name":44,"category_name":97,"show_sort_weight":98,"slug":99},"Literature",80,"literature",{"id":51,"doc_module":4,"doc_module_name":44,"category_name":101,"show_sort_weight":102,"slug":103},"Exam",70,"exam",{"id":105,"doc_module":4,"doc_module_name":44,"category_name":106,"show_sort_weight":107,"slug":108},5,"Comic",60,"comic",{"id":110,"doc_module":4,"doc_module_name":44,"category_name":111,"show_sort_weight":112,"slug":113},6,"Technology",50,"technology",{"id":115,"doc_module":4,"doc_module_name":44,"category_name":116,"show_sort_weight":117,"slug":118},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":44,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":44,"category_name":124,"show_sort_weight":27,"slug":125},9,"Religion & Spirituality","religion-spirituality",{"id":27,"doc_module":4,"doc_module_name":44,"category_name":127,"show_sort_weight":27,"slug":128},"World Cup","world-cup",{"id":130,"doc_module":4,"doc_module_name":44,"category_name":131,"show_sort_weight":130,"slug":132},10,"Lifestyle","lifestyle",{"id":134,"doc_module":4,"doc_module_name":44,"category_name":135,"show_sort_weight":105,"slug":136},19,"General","general"]