[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-125103-en":3,"doc-seo-125103-105":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},125103,7971461740886,"Theodore","https://ap-avatar.wpscdn.com/davatar_3d24733baf745e90a7e4bdd5f77d97b2",8,"Research & Report","MACE - A Machine learning Approach to Chemistry Emulation - Proof-of-concept study","Astrochemical chemistry and dynamics are tightly coupled, requiring a 3D treatment, yet direct chemical-kinetics calculations inside 3D hydro simulations are computationally prohibitive for practical parameter studies. To enable feasible 3D hydro-chemical modeling, MACE replaces the classical chemical approach with a machine-learning emulator. Trained on AGB outflow conditions, it combines an autoencoder for dimensionality reduction with latent ordinary differential equations for temporal evolution, reproducing full chemical pathways and improving performance by 26x on average.","arXiv :2405 .03274v1 [physics .comp-ph] 6 May 2024  \nDraft version May 7, 2024  \nTypeset using LATEX twocolumn style in AASTeX631  \nMACE: A Machine learning Approach to Chemistry Emulation  \nSilke Maes  ,1 Frederik De Ceuster  ,1 Marie Van de Sande  ,2, 3 and Leen Decin 1, 4  \n1 Institute of Astronomy, KU Leuven, Celestijnenlaan 200D, B-3001 Leuven, Belgium  \n2 Leiden Observatory, Leiden University, PO Box 9513, 2300 RA Leiden, The Netherlands  \n3 School of Physics and Astronomy, University of Leeds, Leeds LS2 9JT, United Kingdom  \n4 School of Chemistry, University of Leeds, Leeds LS2 9JT, United Kingdom  \nABSTRACT  \nThe chemistry of an astrophysical environment is closely coupled to its dynamics, the latter often found to be complex. Hence, to properly model these environments a 3D context is necessary. However, solving chemical kinetics within a 3D hydro simulation is computationally infeasible for a even a modest parameter study. In order to develop a feasible 3D hydro-chemical simulation, the classical chemical approach needs to be replaced by a faster alternative. We present mace, a Machine learning Approach to Chemistry Emulation, as a proof-of-concept work on emulating chemistry in a dynamical environment. Using the context of AGB outflows, we have developed an architecture that combines the use of an autoencoder (to reduce the dimensionality of the chemical network) and a set of latent ordinary differential equations (that are solved to perform the temporal evolution of the reduced features) . Training this architecture with an integrated scheme makes it possible to successfully reproduce a full chemical pathway in a dynamical environment. mace outperforms its classical analogue on average by a factor 26 . Furthermore, its efficient implementation in PyTorch results in a sub-linear scaling with respect to the number of hydrodynamical simulation particles.  \nKeywords: Astrochemistry (75) – Computational methods (1965) – Astronomy software (1855) – Chemical reaction network models (2237)– Asymptotic giant branch stars (2100)– Stellar winds (1636)  \n1. INTRODUCTION  \nAstrochemistry, the study of chemistry in space, is a powerful tool. Combining observations of chemical species in astrophysical objects with theoretical predictions allows us to study the physical conditions, as well as to estimate its chemical composition and evolution. Astrochemistry labs are found indifferent environments ranging from dark clouds and protoplanetary disks to different phases of interstellar medium (ISM), and cluster formation in galaxies.  \nThe astrophysical environment impacts its chemistry, and vice versa. For instance, cooling and heating processes as a result of chemical reactions will influence the dynamics. Hence, a hydrodynamics model needs to be coupled with a chemistry model, apart from  \nCorresponding author: Silke Maes  \n[silke.maes@kuleuven.be](silke.maes@kuleuven.be)  \nradiation, in order to fully simulate an astrophysical environment. Moreover, this dynamics is often complex and therefore requires a 3-dimensional approach when modelling it. 3D hydrodynamical modelling is a notoriously computationally expensive process, both ina particle-based and grid-based approach. It is often the case that coupling such hydrodynamics with a classical chemical model in every time step makes it computationally infeasible to explore even a modest physical parameter space. Various research groups have already made elaborate efforts in integrating (limited) chemistry in hydrodynamical simulations. To name a few, Glover & Mac Low (2007a,b); Walch et al. (2015), and Hu et al. (2021) combine hydro and chemistry, amongst other processes, in the case of molecular clouds and the ISM, Lah´en et al. (2020) incorporated both constituents in star formation simulations, Yoneda et al. (2016) and Young et al. (2021) did so for protoplanetary disk research (the latter doing chemistry in a post-processing step), and Richings & Schaye (2016)  \n2  \nfor galaxy f","cbCaik6EUXD5qxjr","https://ap.wps.com/l/cbCaik6EUXD5qxjr","pdf",5008720,1,22,"English","en",105,"# Introduction\n## Motivation: coupling chemistry and 3D hydrodynamics\n## Target environment: AGB circumstellar envelopes\n# Approach: MACE for chemistry emulation\n## Autoencoder-based dimensionality reduction\n## Latent ODEs for temporal evolution\n# Results and performance\n## Reproducing full chemical pathways\n## Scaling and implementation details","[{\"question\":\"Why is 3D hydro-chemical simulation difficult with classical chemistry models?\",\"answer\":\"Because solving chemical kinetics at every time step inside 3D hydrodynamical simulations is computationally infeasible for even modest parameter studies.\"},{\"question\":\"What is the core idea behind MACE?\",\"answer\":\"MACE emulates chemistry in a dynamical environment using machine learning, replacing the classical chemical kinetics approach with a learned surrogate.\"},{\"question\":\"How does MACE model chemical evolution over time in 3D simulations?\",\"answer\":\"It uses an autoencoder to reduce the dimensionality of the chemical network, then uses a set of latent ordinary differential equations to evolve the reduced features temporally.\"}]","MACE - A Machine learning Approach to Chemistry Emulation - Proof-of-concept study | PDF",1785896664,55,{"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},"mace-a-machine-learning-approach-to-chemistry-emulation-proof-of-concept-study","",{"@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/mace-a-machine-learning-approach-to-chemistry-emulation-proof-of-concept-study/125103/",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","2026-08-05",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},"Why is 3D hydro-chemical simulation difficult with classical chemistry models?","Question",{"text":75,"@type":76},"Because solving chemical kinetics at every time step inside 3D hydrodynamical simulations is computationally infeasible for even modest parameter studies.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What is the core idea behind MACE?",{"text":80,"@type":76},"MACE emulates chemistry in a dynamical environment using machine learning, replacing the classical chemical kinetics approach with a learned surrogate.",{"name":82,"@type":73,"acceptedAnswer":83},"How does MACE model chemical evolution over time in 3D simulations?",{"text":84,"@type":76},"It uses an autoencoder to reduce the dimensionality of the chemical network, then uses a set of latent ordinary differential equations to evolve the reduced features temporally.","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"]