[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-122636-en":3,"doc-seo-122636-105":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},122636,1099513958762,"Logic","https://ap-avatar.wpscdn.com/avatar/1000023916a998db790?x-image-process=image/resize,m_fixed,w_180,h_180&k=1784791008015729253",8,"Research & Report","Addressing computational challenges in physical system simulations with machine learning","Machine learning-based data generation is proposed to help researchers study physical systems and processes using computational simulations. High computational cost and the resulting limited datasets restrict analysis and insight. The framework uses a two-step workflow: a supervised predictive surrogate is trained from a small simulated dataset to predict simulation outcomes, then a reinforcement learning agent is trained to generate simulation-like data using guidance from the supervised model. The approach enables efficient exploration of parameter space and deeper understanding without repeated expensive simulations, demonstrated via earthquake rupture physics and new material development case studies.","arXiv :2305 .09627v1 [ cs .LG] 16 May 2023  \nAddressing computational challenges in physical system simulations with machine  \nlearning  \nSabber Ahamed 1a;􀀃, Md Mesbah Uddin 1b;􀀃  \n1a;􀀃sabbers@gmail . com, 1b;􀀃Mesbahuddin1991@gmail . com  \nMay 17, 2023  \nAbstract  \nIn this paper, we present a machine learning-based data generator framework, tailored to aid researchers who utilize simulations to examine various physical systems or processes. High computational costs and the resulting limited data often pose signi􀀌cant challenges to gaining insights into these systems or processes. Our approach involves a two-step process: initially, we train a supervised predictive model using a limited simulated dataset to predict simulation outcome. Subsequently, a reinforcement learning agent is trained to generate accurate, simulation-like data by leveraging the supervised model. With this framework, researchers can generate more accurate data and know the outcomes without running high computational simulations. Which enables them to explore the parameter space more e􀀎ciently and gain deeper insights into physical systems or processes. In this paper, we demonstrate the e􀀋ectiveness of the proposed framework by applying it to two case studies one focusing on earthquake rupture physics and the other on new material development.  \n􀀃 These authors contributed equally to this work.  \n1 Introduction  \nUnderstanding physical systems or processes often involves the use of computational simulations, a method that can be both time-consuming and computationally expensive. For instance, simulating the weather and climate dynamics [1], modeling complex biological systems like protein folding [2], or even understaing earthquake rupture process [3] or seismic activities [4, 5] requires a signi􀀌cant computational demands.These constraints can severely limit the amount of data available for research and impact our ability to gain deep insights into these systems or processes.  \nIn response to these challenges, we propose a novel machine learningbased generator framework designed to enhance the e􀀎ciency of data generation in studies that rely on simulations. This framework operates in a two-step approach. Initially, a supervised predictive model is trained using a small-scale simulated dataset with varying input parameters. This predictive model, functioning as a surrogate for the original physical simulations, is then used to train a reinforcement learning (RL) agent [6] . The RL agent, guided by the feedback from the predictive model, learns to generate more accurate, simulation-like data.  \nThe key advantage of this framework is that it facilitates the generation of larger quantities of data without the need for further high computational simulations. This allows researchers to explore the parameter space more e􀀎 -ciently, thereby gaining deeper insights into the physical systems or processes under study. Furthermore, by employing a reinforcement learning approach, our framework can continually adapt and improve over time, o􀀋ering the potential for even greater accuracy and e􀀎ciency in future data generation.  \nIn this paper, we demonstrate the e􀀋ectiveness of the proposed framework by applying it to two case studies in material science and geodynamics, thereby highlighting its versatility and broad applicability across various domains of physical sciences.  \n2 Generator Framework  \nThe proposed machine learning framework consists of two main components:  \n(1) a supervised predictive model, (2) a reinforcement learning agent. The framework is consists of these interdependent components that work together to generate more simulations like data and predict the outcomes of simula-  \ntions.  \n2.1 Overview  \nFigure 1: The schematic diagram shows how the framework works  \nFirst, we generate some simulated dataset with varying input parameters (Xs) . We then train a supervised predictive model, f (Xs ) to predict the simulation outcomes (ys) . In the second step","cbCaiewAC2wjvdmU","https://ap.wps.com/l/cbCaiewAC2wjvdmU","pdf",2273822,1,15,"English","en",105,"# Abstract\n# 1 Introduction\n# 2 Generator Framework\n## 2.1 Overview\n## 2.2 Supervised Model: Predicting the Outcome of Simulations\n## 2.3 RL Agent: Generate simulation like data","[{\"question\":\"What computational problem does the proposed framework address?\",\"answer\":\"The framework targets high computational costs that limit the amount of data researchers can obtain from simulation-based studies, reducing insight into physical systems or processes.\"},{\"question\":\"How does the framework generate simulation-like data?\",\"answer\":\"It first trains a supervised predictive surrogate on a small simulated dataset, then trains a reinforcement learning agent to generate accurate simulation-like data using the surrogate’s feedback.\"},{\"question\":\"What are the main case studies used to demonstrate effectiveness?\",\"answer\":\"The framework is evaluated with two case studies: earthquake rupture physics and new material development.\"}]","Addressing computational challenges in physical system simulations with machine learning | PDF",1785811845,38,{"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},"addressing-computational-challenges-in-physical-system-simulations-with-machine-learning","",{"@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/addressing-computational-challenges-in-physical-system-simulations-with-machine-learning/122636/",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","2026-08-04",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 computational problem does the proposed framework address?","Question",{"text":75,"@type":76},"The framework targets high computational costs that limit the amount of data researchers can obtain from simulation-based studies, reducing insight into physical systems or processes.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does the framework generate simulation-like data?",{"text":80,"@type":76},"It first trains a supervised predictive surrogate on a small simulated dataset, then trains a reinforcement learning agent to generate accurate simulation-like data using the surrogate’s feedback.",{"name":82,"@type":73,"acceptedAnswer":83},"What are the main case studies used to demonstrate effectiveness?",{"text":84,"@type":76},"The framework is evaluated with two case studies: earthquake rupture physics and new material development.","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},[93,97,101,105,110,115,120,123,128,131,135],{"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":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":106,"slug":138},19,"General","general"]