[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-120085-en":3,"doc-seo-120085-105":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},120085,2336464648322,"Aria","https://ap-avatar.wpscdn.com/avatar/2200025388227c56fec?_k=1778556882303663488",8,"Research & Report","Opportunities for machine learning in scientific discovery","Advances in computing and the scale of available data make machine learning (ML) techniques increasingly applicable across scientific domains. Yet using ML to obtain fundamental, formalized understanding of natural processes remains at an early stage. This review examines how researchers can leverage ML to drive scientific discovery, noting that the opportunity depends on prior knowledge of governing equations and physical properties. It highlights open challenges while showing that principled ML use helps manage complex observational data previously difficult for classic analysis.","arXiv :2405 .04161v1 [ cs .LG] 7 May 2024  \nOpportunities for machine learning in scientific discovery  \nRicardo Vinuesa1,2*, Jean Rabault3 , Hossein Azizpour4,2 , Stefan Bauer5,6 , Bingni W. Brunton7 , Arne Elofsson8,2 , Elias Jarlebring9,2 , Hedvig Kjellstrm4,2 , Stefano Markidis10,2 , David Marlevi11,12 , Paola Cinnella13 , and Steven L. Brunton14  \n1 FLOW, Engineering Mechanics, KTH Royal Institute of Technology, Stockholm, Sweden  \n2 Swedish e-Science Research Centre,(SeRC), Stockholm, Sweden  \n3 IT Department, Norwegian Meteorological Institute, 0313 Oslo, Norway  \n4 Robotics, Perception and Learning, KTH Royal Institute of Technology, Stockholm, Sweden  \n5TUM School of Computation, Information and Technology, Technical University Munich, Munich, Germany  \n6 Helmholtz AI, Helmholtz Center Munich, Munich, Germany  \n7 Department of Biology, University of Washington, Seattle, WA 98195, USA  \n8 Dept. of Biochemistry and Biophysics and Science for Life Laboratory, Stockholm University, 171 21 Solna  \n9 Dept. Mathematics, KTH Royal Institute of Technology, 100 44 Stockholm, Sweden  \n10 Department of Computer Science, KTH Royal Institute of Technology, Stockholm, Sweden  \n11 Dept. Molecular Medicine and Surgery, Karolinska Institutet, 171 77 Stockholm, Sweden  \n12 Inst. for Medical Engineering and Science, Massachusetts Institute of Technology, Cambridge, MA 02139, USA 13 Institut Jean le Rond D’Alembert, Sorbonne Universit, France  \n14 Department of Mechanical Engineering, University of Washington, Seattle, WA 98195, USA  \n*E-mail for correspondence: [rvinuesa@mech.kth.se](rvinuesa@mech.kth.se)  \nABSTRACT  \nTechnological advancements have substantially increased computational power and data availability, enabling the application of powerful machine-learning (ML) techniques across various fields. However, our ability to leverage ML methods for scientific discovery, [i.e.](i.e. to)[ to](i.e. to) obtain fundamental and formalized knowledge about natural processes, is still in its infancy. In this review, we explore how the scientific community can increasingly leverage ML techniques to achieve scientific discoveries. We observe that the applicability and opportunity of ML depends strongly on the nature of the problem domain, and whether we have full (e.g., turbulence), partial (e.g., computational biochemistry), or no (e.g., neuroscience) a-priori knowledge about the governing equations and physical properties of the system. Although challenges remain, principled use of ML is opening up new avenues for fundamental scientific discoveries. Throughout these diverse fields, there is a theme that ML is enabling researchers to embrace complexity in observational data that was previously intractable to classic analysis and numerical investigations. Keywords: machine learning (ML); deep learning (DL); artificial intelligence (AI); scientific discovery; physics; life sciences; computer science  \nIntroduction  \nMachine learning (ML) has shown great potential to transform a broad range of domains [1–5], and it is increasingly being applied to problems in science and engineering. ML has been widely used for predictive tasks in these areas, and despite an initial promising phase where ML methods have outperformed well-established techniques [6– 10], such predictive applications are starting to exhibit diminishing returns. There are, by contrast, increasing opportunities in the academic usage of ML for scientific discovery, i.e. answering challenging scientific questions while leveraging existing fundamental knowledge. Such focus on scientific discovery can move the frontiers of science forward when progress in more traditional methods has slowed. Furthermore, the development of novel and more powerful ML methods can help to tackle some open subjects in the context of predictions from scarce, noisy, or incomplete data, out-of-sample generalization, extreme-event predictions and predictions under uncertainty.  \nScience is fundamentally interest","cbCaiujsjUE41D4e","https://ap.wps.com/l/cbCaiujsjUE41D4e","pdf",5617750,1,22,"English","en",105,"# Abstract\n# Introduction\n## ML for predictive tasks and diminishing returns\n## ML for scientific discovery with existing knowledge\n## Complexity in natural sciences","[{\"question\":\"What is the main focus of the review on machine learning and scientific discovery?\",\"answer\":\"The review explores how the scientific community can leverage machine learning techniques to achieve scientific discoveries, emphasizing the shift from predictive uses toward discovery-oriented questions informed by existing knowledge.\"},{\"question\":\"Why are opportunities for ML in scientific discovery strongly dependent on the problem domain?\",\"answer\":\"Applicability depends on whether researchers have full, partial, or no prior knowledge about governing equations and physical properties, which changes how ML can be used effectively.\"},{\"question\":\"What challenges and limitations are discussed regarding using ML for discovery?\",\"answer\":\"The text notes remaining challenges while arguing that principled ML use is opening new avenues, especially for dealing with complexity in observational data that is hard for classic analysis and numerical investigations.\"}]","Opportunities for machine learning in scientific discovery | PDF",1785728081,55,{"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},"opportunities-for-machine-learning-in-scientific-discovery","",{"@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/opportunities-for-machine-learning-in-scientific-discovery/120085/",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","2026-08-03",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 is the main focus of the review on machine learning and scientific discovery?","Question",{"text":75,"@type":76},"The review explores how the scientific community can leverage machine learning techniques to achieve scientific discoveries, emphasizing the shift from predictive uses toward discovery-oriented questions informed by existing knowledge.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"Why are opportunities for ML in scientific discovery strongly dependent on the problem domain?",{"text":80,"@type":76},"Applicability depends on whether researchers have full, partial, or no prior knowledge about governing equations and physical properties, which changes how ML can be used effectively.",{"name":82,"@type":73,"acceptedAnswer":83},"What challenges and limitations are discussed regarding using ML for discovery?",{"text":84,"@type":76},"The text notes remaining challenges while arguing that principled ML use is opening new avenues, especially for dealing with complexity in observational data that is hard for classic analysis and numerical investigations.","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"]