[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-117237-en":3,"doc-seo-117237-105":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},117237,2336464648322,"Aria","https://ap-avatar.wpscdn.com/avatar/2200025388227c56fec?_k=1778556882303663488",8,"Research & Report","Uncertainty-Informed Model Selection Method for Nonlinear System Identification and Interpretable Machine Learning - Paper","Modeling uncertainty plays a central role in data-driven modeling and machine learning, yet it complicates selecting the best model among many candidates. An uncertainty-informed method is proposed to directly address the model selection challenge under uncertainty. Experiments use a dataset generated from a complex system model, showing clear effectiveness and improved performance over conventional approaches. The approach requires minimal training-data length and flexible model-type assumptions, enabling use across diverse modeling frameworks.","Final accepted manuscript  \n2024 32nd Mediterranean Conference on Control and Automation (MED)  \nJune 11-14, 2024 | Chania, Crete, Greece. pp. 909-914.  \nUncertainty-Informed Model Selection Method for Nonlinear System Identification and Interpretable Machine Learning*  \nYuanlin Gu, Hua-Liang Wei  \nAbstract—Modeling uncertainty has been an active and important topic in the fields of data-driven modeling and machine learning. Uncertainty ubiquitously exists in any data modeling process, making it challenging to identify the optimal models among many potential candidates. This article proposesan uncertainty-informed method to address the model selection problem. The performance of the proposed method is evaluated on a dataset generated from a complex system model. The experimental results demonstrate the effectiveness of the proposed method and its superiority over conventional approaches. This method has minimal requirements for the length of training data and model types, making it applicable for various modeling frameworks.  \nI. INTRODUCTION  \nThe typical process of data-driven modeling involves several stages including data collection, preprocessing, model training and model validation. For most data-driven modeling methods, it is often necessary to define some training parameters prior to model training. For instance, when building a neural network model, parameters such as the number of epochs, the estimation/optimization algorithm, evaluation metrics, and network structure (including layer types, number of layers, and neurons per layer) must be specified first [1], [2] . For some regression-based models, such as the Nonlinear AutoRegressive Moving Average with eXogenous inputs (NARMAX) model [3], [4], the determination of model structure and the generation of candidate linear and nonlinear model terms are essential. The following processes such as the selection of important model terms, the determination of model size (i.e., the number of terms), model validity test, and model performance evaluation plays an important and central role [5] . Due to these factors, for a given data modeling task, there could be a vast number of candidate solutions. In practice it is always challenging to effectively identify the best model or the best set of models among these candidates.  \nNumerous model selection and model size determination methods have been introduced and widely applied in data modeling, system identification, and machine learning. Among the popular ones are the Akaike Information Criterion (AIC)  \n[6] and Bayesian Information Criterion (BIC) [7] . These criteria aim to find a balance between model performance and  \n*Research supported by STFC (Ref. ST-Y001524-1), NERC (Ref. NE/V002511/1), and the University of Stirling.  \nYuanlin Gu is with the Division of Computer Science and Maths, University of Stirling, Stirling, FK9 4LA, United Kingdom. (e-mail [address: yuanlin.gu@stir.ac.uk](address: yuanlin.gu@stir.ac.uk)).  \nHua-Liang Wei is with the Department of Automatic Control and Systems Engineering, University of Sheffield, Sheffield, S1 3JD (*Corresponding author phone: 0044 1142225198; e-mail: [w.hualiang@sheffield.ac.uk](w.hualiang@sheffield.ac.uk)).  \ncomplexity by incorporating measures for both. They have found extensive use across various applications [8], [9], [10] . Additionally, methods such as prediction error sum of squares (APRESS) [11] were developed to address model selection challenges in complex nonlinear system identification, which have been proven effective in diverse fields. Cross-validation is another widely utilized technique, particularly suitable for black-box models such as neural networks [12], [13] .  \nIn recent years, uncertainty analysis has become an important and hot topic in data modeling fields [14], [15] . When strong uncertainties exist, models can become unreliable, particularly when predicting peak values in some specific applications, e.g., space weather prediction [16] . This can i","cbCaibFyMwjAJHxP","https://ap.wps.com/l/cbCaibFyMwjAJHxP","pdf",530786,1,6,"English","en",105,"# I. INTRODUCTION\n## Data-driven modeling workflow and model selection challenges\n## Common criteria: AIC, BIC, APRESS, cross-validation\n## Uncertainty analysis and prediction uncertainty quantification","[{\"question\":\"What problem does the proposed method target?\",\"answer\":\"It targets uncertainty-related model selection, where many candidate models exist and it is difficult to identify the best and most representative ones while requiring uncertainty quantification during modeling.\"},{\"question\":\"How is performance evaluated in the article?\",\"answer\":\"Performance is evaluated on a dataset generated from a complex system model, and the results demonstrate the method’s effectiveness and superiority over conventional approaches.\"},{\"question\":\"What practical advantages does the method offer?\",\"answer\":\"The method has minimal requirements for training data length and supports different model types, making it applicable to various modeling frameworks.\"}]","Uncertainty-Informed Model Selection Method for Nonlinear System Identification and Interpretable Machine Learning - Paper | PDF",1785674604,15,{"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},"uncertainty-informed-model-selection-method-for-nonlinear-system-identification-and-interpretable-machine-learning-paper","",{"@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/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/uncertainty-informed-model-selection-method-for-nonlinear-system-identification-and-interpretable-machine-learning-paper/117237/",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","2026-08-02",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 the proposed method target?","Question",{"text":75,"@type":76},"It targets uncertainty-related model selection, where many candidate models exist and it is difficult to identify the best and most representative ones while requiring uncertainty quantification during modeling.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How is performance evaluated in the article?",{"text":80,"@type":76},"Performance is evaluated on a dataset generated from a complex system model, and the results demonstrate the method’s effectiveness and superiority over conventional approaches.",{"name":82,"@type":73,"acceptedAnswer":83},"What practical advantages does the method offer?",{"text":84,"@type":76},"The method has minimal requirements for training data length and supports different model types, making it applicable to various modeling frameworks.","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},[93,97,101,105,110,114,119,122,127,130,134],{"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":21,"doc_module":4,"doc_module_name":46,"category_name":111,"show_sort_weight":112,"slug":113},"Technology",50,"technology",{"id":115,"doc_module":4,"doc_module_name":46,"category_name":116,"show_sort_weight":117,"slug":118},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":46,"category_name":124,"show_sort_weight":125,"slug":126},9,"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"]