[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-124870-en":3,"doc-seo-124870-105":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},124870,549758252649,"Ivy","https://ap-avatar.wpscdn.com/avatar/8000253669c5317157?_k=1778319167496531819",6,"Technology","Holonic Learning - A Flexible Agent-based Distributed Machine Learning Framework","Ever-increasing data and computing resources have driven machine learning toward distributed paradigms to improve scalability while addressing privacy and security needs. The work presents Holonic Learning (HoL), a collaborative, privacy-focused learning framework for training deep models with holonic concepts. HoL organizes learning into a structured self-similar hierarchy using holon-level aggregation, intra-holon commitment, and communication patterns. It implements HoloAvg for weighted aggregation and validates convergence on IID and Non-IID MNIST experiments, analyzing behaviors across hierarchical designs and data distributions. Results show competitive performance, especially for Non-IID data.","Holonic Learning: A Flexible Agent-based Distributed Machine  \nLearning Framework  \nAhmad Esmaeili  \nPurdue University West Lafayette, United States [aesmaei@purdue.edu](aesmaei@purdue.edu)  \nZahra Ghorrati  \nPurdue University West Lafayette, United States [zghorrat@purdue.edu](zghorrat@purdue.edu)  \nEric T. Matson  \nPurdue University West Lafayette, United States [ematson@purdue.edu](ematson@purdue.edu)  \narXiv :2401 . 10839v1 [ cs .DC] 29 Dec 2023  \nABSTRACT  \nEver-increasing ubiquity of data and computational resources in the last decade have propelled a notable transition in the machine learning paradigm towards more distributed approaches. Such a transition seeks to not only tackle the scalability and resource distribution challenges but also to address pressing privacy and security concerns. To contribute to the ongoing discourse, this paper introduces Holonic Learning (HoL), a collaborative and privacy-focused learning framework designed for training deep learning models. By leveraging holonic concepts, the HoL framework establishes a structured self-similar hierarchy in the learning process, enabling more nuanced control over collaborations through the individual model aggregation approach of each holon, along with their intra-holon commitment and communication patterns. HoL, in its general form, provides extensive design and flexibility potentials. For empirical analysis and to demonstrate its effectiveness, this paper implements HoloAvg, a special variant of HoL that employs weighted averaging for model aggregation across all holons. The convergence of the proposed method is validated through experiments on both IIDand Non-IID settings ofthe standard MNISt dataset. Furthermore, the performance behaviors of HoL are investigated under various holarchical designs and data distribution scenarios. The presented results affirm HoL’s prowess in delivering competitive performance particularly, in the context of the Non-IID data distribution.  \nKEYWORDS  \nDistributed learning, Holonic Learning, Collaborative Learning, Edge Computing  \nACM Reference Format:  \nAhmad Esmaeili, Zahra Ghorrati, andEricT. Matson. 2024. Holonic Learning: A Flexible Agent-based Distributed Machine Learning Framework. In Proc. of the 23rd International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2024), Auckland, New Zealand, May 6 – 10, 2024, IFAAMAS, 9 pages.  \n1 INTRODUCTION  \nToday’s interconnected world, marked by the proliferation of information and computational capabilities across vast digital landscapes, has led to critical challenges concerning the scalability, adaptability, speed, and support for diversity in the traditional centralized learning models. As a solution to such limitations and harnessing the collective power of the networked devices—whether servers or edge devices—recent decade has witnessed a substantial  \nProc. of the 23rd International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2024), N. Alechina, V. Dignum, M. Dastani, J.S. Sichman (eds.), May 6 – 10, 2024, Auckland, New Zealand. © 2024 International Foundation for Autonomous Agents and Multiagent Systems ([www.ifaamas.org](www.ifaamas.org)). This work is licenced under the Creative Commons Attribution 4 .0 International (CC-BY 4 .0) licence.  \ngrowth in the rise of distributed learning methodologies that not only expedite training process but also enhances robustness and privacy.  \nFederated Learning (FL) [17] stands out as one of the prominent distributed learning approaches attracting attentions from both academia and industry [26] . By empowering end users to participate in the collaborative learning process while retaining ownership of their data, FL ensures both data security and privacy on one hand, and leveraging the potential of computationally constrained devices on the other. Despite its apparent simplicity in its original format, FL introduces a series of intricate open problems and challenges tackled by a plethora of researc","cbCairKDzsicfaRl","https://ap.wps.com/l/cbCairKDzsicfaRl","pdf",1170045,1,9,"English","en",105,"# Abstract\n# Introduction\n## Federated Learning and open challenges\n## Related decentralized and relaxed-assumption approaches\n## Hierarchical federated learning background","[{\"question\":\"What problem does Holonic Learning (HoL) aim to address?\",\"answer\":\"HoL targets scalability and resource-distribution limitations of centralized training while also tackling privacy and security concerns in distributed learning.\"},{\"question\":\"How does HoL structure collaboration during training?\",\"answer\":\"HoL builds a structured self-similar hierarchy of holons, where collaboration is controlled through holon-level aggregation along with intra-holon commitment and communication patterns.\"},{\"question\":\"What is HoloAvg and how is it used in the experiments?\",\"answer\":\"HoloAvg is a special variant of HoL that performs weighted averaging for model aggregation across holons. The framework is validated through experiments on IID and Non-IID settings of MNIST and compared across hierarchical and data distribution scenarios.\"}]","Holonic Learning - A Flexible Agent-based Distributed Machine Learning Framework | PDF",1785895132,23,{"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},"holonic-learning-a-flexible-agent-based-distributed-machine-learning-framework","",{"@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/technology/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/holonic-learning-a-flexible-agent-based-distributed-machine-learning-framework/124870/",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","2026-08-05",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 problem does Holonic Learning (HoL) aim to address?","Question",{"text":75,"@type":76},"HoL targets scalability and resource-distribution limitations of centralized training while also tackling privacy and security concerns in distributed learning.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does HoL structure collaboration during training?",{"text":80,"@type":76},"HoL builds a structured self-similar hierarchy of holons, where collaboration is controlled through holon-level aggregation along with intra-holon commitment and communication patterns.",{"name":82,"@type":73,"acceptedAnswer":83},"What is HoloAvg and how is it used in the experiments?",{"text":84,"@type":76},"HoloAvg is a special variant of HoL that performs weighted averaging for model aggregation across holons. The framework is validated through experiments on IID and Non-IID settings of MNIST and compared across hierarchical and data distribution scenarios.","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},[93,97,101,105,110,113,118,123,127,130,134],{"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":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":111,"slug":112},50,"technology",{"id":114,"doc_module":4,"doc_module_name":46,"category_name":115,"show_sort_weight":116,"slug":117},7,"Healthcare",40,"healthcare",{"id":119,"doc_module":4,"doc_module_name":46,"category_name":120,"show_sort_weight":121,"slug":122},8,"Research & Report",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"]