[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-120368-en":3,"doc-seo-120368-105":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":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":26,"seo_description":14,"update_tm":27,"read_time":28},120368,687197207057,"Sage","https://ap-avatar.wpscdn.com/davatar_29158cc5080c5b710cf443261637dec0",8,"Research & Report","A Combined Transfer Learning Physics Informed Interpretable Machine Learning Approach to Modelling the Shear Strength of Concrete Walls","Data scarcity constrains many civil engineering modelling tasks, particularly for emerging materials and uncommon structural configurations. This study proposes an interpretable machine learning framework that combines Transfer Learning with Physics-Informed Machine Learning, using Genetic Expression Programming. Domain physics from ACI shear strength equations is incorporated through feature engineering, while the 2M2P algorithm transfers knowledge from common wall geometries to less common ones under limited data. Results indicate TL-PIML outperforming both traditional data-driven models and existing code equations, especially in severe data-constrained settings.","A Combined Transfer Learning Physics Informed Interpretable Machine Learning Approach to Modelling the Shear Strength of Concrete Walls  \nJacob Dylan Murphy 1, Stephanie German Paal 1  \n1Texas A&M University, United States of America  \n[jacob70920@tamu.edu](jacob70920@tamu.edu)  \nABSTRACT  \nData scarcity is a persistent challenge in civil engineering, especially for emerging materials and uncommon structural configurations. This study presents a novel, interpretable machine learning framework combining Transfer Learning (TL) and Physics-Informed Machine Learning (PIML) using Genetic Expression Programming (GEP) . Physical knowledge, based on ACI shear strength equations, is injected through feature engineering, while the 2M2P algorithm enables knowledge transfer from common wall geometries to less common ones with limited data. Results show that the TL-PIML models often outperform both traditional data-driven models and existing code equations, particularly under severe data constraints. The approach yields interpretable, physically consistent models that achieve higher utilizing limited training data.  \nThis work demonstrates the effectiveness of combining TL and PIML for improving predictive modelling in structural engineering and offers a robust  \nstrategy for extending machine learning applications to data-sparse scenarios.  \nKEYWORDS  \nMachine Learning, Transfer Learning, Physis Informed Machine Learning,  \nReinforced Concrete Walls, Shear Strength  \n1. INTRODUCTION  \nCivil engineering is a domain that suffers from a lack of data in many of its different disciplines, especially in regard to emerging materials and methods. The reason for the scarcity of data within civil engineering is the cost and time associated with carrying out the required experimental tests to construct a sufficient database (Pak and Paal 2022, Li, Zhu et al. 2024) . In recent years both transfer learning (TL), and physics-informed machine  \nlearning (PIML) have gained traction and have been successfully employed in a myriad of domains, many of which are related to issues arising from data scarcity.  \nTL offers a promising approach for modelling complex structural behaviour when data is scarce. Inspired by the way humans apply past experiences to new tasks, TL enhances predictive accuracy by transferring knowledge from a related source domain to a target domain (Pan and Yang 2010) .  \nUnlike traditional ML methods that train solely on a single dataset, TL enables models to leverage external knowledge, improving performance even with limited target data. Reducing the resources needed for modelling and analysis lowers the barrier to entry for individuals, institutions, and companies to explore innovative solutions and address engineering challenges. This broader accessibility encourages greater contributions to the understanding and design of civil infrastructure  \nSimilar to TL, PIML is a machine learning method that leverages existing physical knowledge to improve the performance of a model without necessitating an increase in data. Instead of utilizing data driven methods, PIML integrates established physical laws into the training process of machine learning models, offering a hybrid approach that combines data-driven techniques with domain knowledge. Unlike traditional ML methods that rely entirely on data, PIML incorporates physical insights to improve model performance, especially in scenarios with sparse or noisy datasets. This fusion allows PIML to generalize better, reduce overfitting, and maintain robustness in complex systems where conventional ML might struggle.  \nIn civil and structural engineering, where highquality data is often limited due to cost and logistical challenges, PIML offers a practical solution by using physics to compensate for data scarcity (Raissi, Perdikaris, and Karniadakis 2019) . It can be used to adapt to the available data: increasing reliance on physics when data is scarce and vice versa when data is abundant. Thi","cbCaitjiRGQCLWZD","https://ap.wps.com/l/cbCaitjiRGQCLWZD","pdf",1749475,1,"English","en",105,"# Abstract\n# Introduction\n## Data scarcity in civil engineering\n## Transfer learning for limited data\n## Physics-informed machine learning and interpretability\n## Feature engineering with physical laws","[{\"question\":\"What is the main purpose of the proposed framework?\",\"answer\":\"To model the shear strength of concrete walls using a combined Transfer Learning and Physics-Informed Machine Learning approach that remains effective under limited data.\"},{\"question\":\"How is physical knowledge incorporated into the model?\",\"answer\":\"Physical knowledge based on ACI shear strength equations is injected through feature engineering.\"},{\"question\":\"Why does the paper claim the approach is more reliable than purely data-driven models?\",\"answer\":\"The framework transfers knowledge from common wall geometries via 2M2P and embeds physics to improve generalization, robustness, and interpretability, leading to better performance under severe data constraints.\"}]","A Combined Transfer Learning Physics Informed Interpretable Machine Learning Approach to Modelling the Shear Strength of Concrete Walls | PDF",1785729697,20,{"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},"a-combined-transfer-learning-physics-informed-interpretable-machine-learning-approach-to-modelling-the-shear-strength-of-concrete-walls","",{"@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/a-combined-transfer-learning-physics-informed-interpretable-machine-learning-approach-to-modelling-the-shear-strength-of-concrete-walls/120368/",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","2026-08-03",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 is the main purpose of the proposed framework?","Question",{"text":74,"@type":75},"To model the shear strength of concrete walls using a combined Transfer Learning and Physics-Informed Machine Learning approach that remains effective under limited data.","Answer",{"name":77,"@type":72,"acceptedAnswer":78},"How is physical knowledge incorporated into the model?",{"text":79,"@type":75},"Physical knowledge based on ACI shear strength equations is injected through feature engineering.",{"name":81,"@type":72,"acceptedAnswer":82},"Why does the paper claim the approach is more reliable than purely data-driven models?",{"text":83,"@type":75},"The framework transfers knowledge from common wall geometries via 2M2P and embeds physics to improve generalization, robustness, and interpretability, leading to better performance under severe data constraints.","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"]