[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-126714-en":3,"doc-seo-126714-105":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":4,"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},126714,962084925782,"Ava Thompson","https://ap-avatar.wpscdn.com/davatar_9964176cb1d06d4a9deccf72a44ae3dc",8,"Research & Report","IMSRG-Net - A machine learning-based solver for In-Medium Similarity Renormalization Group","IMSRG-Net introduces a machine learning solver for the in-medium Similarity Renormalization Group (IMSRG) flow equation by approximating the Magnus operators Ω(s). The method constructs a physics-informed loss function that embeds the IMSRG flow equation into the neural network model. Trained on a small dataset with flow parameters up to s = 20, covering roughly one-eighth to one-quarter of the full flow, IMSRG-Net achieves high-accuracy extrapolations of ground-state energies and charge radii for 16O and 40Ca, and it can derive effective interactions within a valence space.","IMSRG-Net: A machine learning-based solver for In-Medium Similarity Renormalization Group  \narXiv :2306 .08878v2 [nucl-th] 18 Sep 2023  \nSota Yoshida 1, ∗  \n1 Institute for Promotion of Higher Academic Education,  \nUtsunomiya University, Mine, Utsunomiya, 321-8505, Japan  \n(Dated: September 20, 2023)  \nWe present a novel method, IMSRG-Net, which utilizes machine learning techniques as a solver for the inmedium Similarity Renormalization Group (IMSRG) . The primary objective of IMSRG-Net is to approximate the Magnus operators Ω(s) in the IMSRG flow equation, thereby offering an alternative to the computationally intensive part of IMSRG calculations. The key idea of IMSRG-Net is its design of the loss function inspired by physics-informed neural networks to encode the underlying physics, i.e., IMSRG flow equation, into the model. Through training on a dataset comprising ten data points with flow parameters up to s = 20, capturing approximately one-eighth to one-quarter of the entire flow, IMSRG-Net exhibits remarkable accuracy in extrapolating the ground state energies and charge radii of 16 O and 40 Ca. Furthermore, this model demonstrates effectiveness in deriving effective interactions for a valence space.  \nI. INTRODUCTION  \nThe in-medium Similarity Renormalization Group (IMSRG) method [1–5] is a highly powerful framework to study nuclear many-body systems. This method serves as an ab initio technique for investigating the properties of nuclei near subshell closures, while also enabling the systematic derivation of effective interactions and operators for a valence space. The IMSRG method is formulated by the unitary transformation of operators, such as the Hamiltonian, through the IMSRG flow equation. The objective is to decouple a target subspace from the rest of many-body Hilbert space. For ground state calculations, this entails decoupling particle-hole excitations from the reference state, whereas for deriving effective interactions, the focus is on decoupling the valence space from the core and outside (excluded) space. Notably, recent studies have extended the application of the IMSRG to heavier nuclei, such as 132 Sn and 208Pb [6, 7] .  \nAlthough the IMSRG method is a powerful approach, it is still computationally demanding to perform numerous calculations for different nuclei and input nuclear interactions. Consequently, it is crucial to develop efficient methods for conducting IMSRG calculations. The construction of such emulators or surrogate models has emerged as a prominent research topic within the nuclear physics community, providing as a key tool for comprehending and evaluating the uncertainties associated with nuclear many-body calculations and realistic nuclear potentials. A notable example of such emulators is the eigenvector continuation (EC) method [8–10], which has been extensively applied to diverse nuclear many-body problems [11–21] . Its significance has been recognized from abroader perspective as model order reduction [22, 23] . However, applying the EC method to IMSRG calculations poses challenges since the EC method primarily operates on manybody wave functions (eigenvectors of a Hamiltonian), while IMSRG calculations are performed on many-body operators, such as the Hamiltonian.  \n∗ [syoshida@cc.utsunomiya-u.ac.jp](syoshida@cc.utsunomiya-u.ac.jp)  \nIn this work, we present an alternative approach to constructing a surrogate model for IMSRG, employing a data-driven technique based on machine learning. The nuclear physics community has witnessed diverse applications of machine learning-based models to replicate or assist nuclear many-body calculations (see, e.g., Refs. [24]) . We propose a machinelearning-based solver for the IMSRG flow equation, named IMSRG-Net, which is inspired by the physics-informed neural network (PINN) [25, 26] . The neural network model, IMSRG-Net, is trained to approximate Magnus operators in IMSRG methods as a function of the flow parameter s. The primary objective of t","cbCait5Wb1pdjuri","https://ap.wps.com/l/cbCait5Wb1pdjuri","pdf",1921665,1,"English","en",105,"# Introduction\n## Computational motivation and surrogate/emulator approaches\n# Methodology\n## IMSRG-Net design and physics-informed loss\n## Training setup and computational framework\n# Results\n## Extrapolation accuracy for nuclei and effective interactions\n# Summary","[{\"question\":\"What problem does IMSRG-Net solve in IMSRG calculations?\",\"answer\":\"IMSRG-Net approximates the Magnus operators Ω(s) in the IMSRG flow equation to avoid the high computational cost of traditional IMSRG calculations.\"},{\"question\":\"How is the neural network trained to respect IMSRG physics?\",\"answer\":\"The loss function is inspired by physics-informed neural networks, encoding the IMSRG flow equation so that both Ω and flow-equation-related indicators (derivatives) guide backpropagation.\"},{\"question\":\"Which nuclear observables and nuclei are used to test the method?\",\"answer\":\"The approach is tested on ground-state energies and charge radii of 16O and 40Ca, and on derivation of effective interactions for sd- and pf-shell nuclei.\"}]","IMSRG-Net - A machine learning-based solver for In-Medium Similarity Renormalization Group | PDF",1785934370,20,{"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},"imsrg-net-a-machine-learning-based-solver-for-in-medium-similarity-renormalization-group","",{"@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/imsrg-net-a-machine-learning-based-solver-for-in-medium-similarity-renormalization-group/126714/",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","2026-08-05",true,{"@type":64,"interactionType":65,"userInteractionCount":4},"InteractionCounter",{"@type":66},"ViewAction",{"@type":68,"mainEntity":69},"FAQPage",[70,76,80],{"name":71,"@type":72,"acceptedAnswer":73},"What problem does IMSRG-Net solve in IMSRG calculations?","Question",{"text":74,"@type":75},"IMSRG-Net approximates the Magnus operators Ω(s) in the IMSRG flow equation to avoid the high computational cost of traditional IMSRG calculations.","Answer",{"name":77,"@type":72,"acceptedAnswer":78},"How is the neural network trained to respect IMSRG physics?",{"text":79,"@type":75},"The loss function is inspired by physics-informed neural networks, encoding the IMSRG flow equation so that both Ω and flow-equation-related indicators (derivatives) guide backpropagation.",{"name":81,"@type":72,"acceptedAnswer":82},"Which nuclear observables and nuclei are used to test the method?",{"text":83,"@type":75},"The approach is tested on ground-state energies and charge radii of 16O and 40Ca, and on derivation of effective interactions for sd- and pf-shell nuclei.","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"]