[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-118463-en":3,"doc-seo-118463-105":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},118463,2336464648746,"Skyler","https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c",8,"Research & Report","Inherent structural descriptors via machine learning","Machine learning is proposed to extract a small set of physically meaningful collective variables for complex systems, addressing limits of conventional collective variable choices in simulations. The method associates instantaneous configurations with their zero-temperature inherent structures, using an autoencoder-like neural network trained to map finite-temperature structural descriptors to inherent-structure descriptors. Applied to structural transitions in nanoclusters, the approach analyzes structural complexity for an experimentally relevant 147-atom gold system and enables computation of free-energy landscapes, transition rates, and non-equilibrium melting/freezing behavior.","arXiv :2407 . 17924v1 [physics .comp-ph] 25 Jul 2024  \nInherent structural descriptors via machine learning  \nEmanuele Telari 1 , Antonio Tinti 1,* , Manoj Settem 1 , Morgan Rees2 , Henry Hoddinott2,3 , Malcolm Dearg4 , Bernd von Issendorff5 , Georg  \nHeld3 , Thomas J.A. Slater4 , Richard E. Palmer2 , Luca Maragliano6,7 , Riccardo Ferrando8 , and Alberto Giacomello 1  \n1 Dipartimento di Ingegneria Meccanica e Aerospaziale, Sapienza Universit`a di Roma, Via Eudossiana 18, Roma, 00184, Italy  \n2 Nanomaterials lab, Mechanical Engineering, Swansea University, Bay Campus, Fabian way, Swansea, SA1 8EN, UK  \n3 Diamond Light Source, Harwell Science and Innovation Campus, Fermi Ave, Didcot, OX11 0DE, England  \n4 Cardiff Catalysis Institute, School of Chemistry, Cardiff University, Translational Research Hub, Cardiff, CF24 4HQ, Wales  \n5 Department of Physics, Albert-Ludwigs-Universit¨at, Freiburg in Breisgau, 79098, Germany  \n6 Dipartimento di Scienze della Vita e dell’Ambiente, Universit`a Politecnica delle Marche, Via Brecce Bianche, Ancona, 60131, Italy  \n7 Center for Synaptic Neuroscience and Technology, Istituto Italiano di Tecnologia, Largo Rosanna Benzi 10, Genova, 16132,  \nItaly  \n8 Dipartimento di Fisica, Universit`a di Genova, Via Dodecaneso, Genova, 16146, Italy  \n* Corresponding author: antonio . tinti[at][uniroma1. it](uniroma1. it)  \nJuly 26, 2024  \nAbstract  \nFinding proper collective variables for complex systems and processes is one of the most challenging tasks in simulations [1], which limits the interpretation of experimental and simulated data [2] and the application of enhanced sampling techniques [3, 4] . Here, we propose a machine  \nlearning approach able to distill few, physically relevant variables by associating instantaneous configurations of the system to their corresponding inherent structures as defined in liquids theory [5] . We apply this ap  \nproach to the challenging case of structural transitions in nanoclusters [6], managing to characterize and explore the structural complexity [7]  \nof an experimentally relevant system constituted by 147 gold atoms [8] .  \nOur inherent-structure variables are shown to be effective at computing complex free-energy landscapes, transition rates, and at describing non-equilibrium melting and freezing processes. The effectiveness of this machine learning strategy guided by the generally-applicable concept of inherent structures [5] shows promise to devise collective variables for avast range of systems, including liquids [9], glasses [10], and proteins [11] .  \n1 Introduction  \nDescribing atomic/molecular processes is a notoriously difficult endeavour [1] even in apparently simple cases such as the isomerization of a small molecule [12] . Producing a low-dimensional representation of such processes usually requires the introduction of functions of the system coordinates, called collective variables (CVs) . CVs can be exploited in advanced simulation techniques [13, 4, 14] for accelerated sampling, FE calculations, and identification of transition mechanisms for a variety of phenomena, including transitions in hard [13], soft and biological [12] matter, and chemical reactions. More generally, starting from the unpractical description in terms of atomic coordinates, CVs attempt to distill essential physical information about complex processes including nonequilibrium ones [15] .  \nRecently, machine learning (ML) has emerged as an invaluable tool for the discovery of CVs [16, 17, 18, 19] in overly complicated systems or when physical intuition fails. In this work, we introduce a generally applicable ML approach for characterizing structural transitions of actual physical systems. We define CVs capable of discriminating structural motifs in noisy finite-temperature configurations based on their zero-temperature counterparts, taking inspiration from the inherent structure concept which we borrow from the theory of liquids [5](Fig. 1a) . In order to devise few, physically inf","cbCaickpveoNE2jV","https://ap.wps.com/l/cbCaickpveoNE2jV","pdf",10193603,1,41,"English","en",105,"# Abstract\n# 1 Introduction","[{\"question\":\"What problem does the document address in molecular simulations?\",\"answer\":\"Choosing proper collective variables for complex systems is difficult, which limits interpretation of simulated/experimental data and the use of enhanced sampling methods.\"},{\"question\":\"How does the proposed machine-learning approach define inherent-structure variables?\",\"answer\":\"It trains an autoencoder-style neural network to map structural descriptors from finite-temperature configurations to their corresponding zero-temperature inherent-structure descriptors, producing latent inherent-structure variables (ISVs).\"},{\"question\":\"Where is the approach applied and what outcomes are reported?\",\"answer\":\"The method is applied to structural transitions in metal nanoclusters, including an experimentally relevant 147-atom gold system, enabling free-energy landscape computation, transition-rate estimation, and analysis of non-equilibrium melting and freezing processes.\"}]","Inherent structural descriptors via machine learning | PDF",1785683734,103,{"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},"inherent-structural-descriptors-via-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/inherent-structural-descriptors-via-machine-learning/118463/",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 document address in molecular simulations?","Question",{"text":75,"@type":76},"Choosing proper collective variables for complex systems is difficult, which limits interpretation of simulated/experimental data and the use of enhanced sampling methods.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does the proposed machine-learning approach define inherent-structure variables?",{"text":80,"@type":76},"It trains an autoencoder-style neural network to map structural descriptors from finite-temperature configurations to their corresponding zero-temperature inherent-structure descriptors, producing latent inherent-structure variables (ISVs).",{"name":82,"@type":73,"acceptedAnswer":83},"Where is the approach applied and what outcomes are reported?",{"text":84,"@type":76},"The method is applied to structural transitions in metal nanoclusters, including an experimentally relevant 147-atom gold system, enabling free-energy landscape computation, transition-rate estimation, and analysis of non-equilibrium melting and freezing processes.","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"]