[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-125606-en":3,"doc-seo-125606-105":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},125606,549758146520,"Patrick","https://ap-avatar.wpscdn.com/avatar/80002397d8c0411e94?_k=1775819394049821470",8,"Research & Report","Robustness of Local Predictions in Atomistic Machine Learning Models - Paper","Atomistic machine learning models for molecules and materials often express a global target as sums of atom-centered, local contributions, enabling linear-scaling simulations and interpretable post-hoc analysis. Yet only the global quantity is rigorously defined, leaving uncertainty about how trustworthy the predicted local terms are. The work introduces a quantitative metric called local prediction rigidity (LPR) to evaluate robustness of locally decomposed predictions, studies its dependence on model training and training dataset composition, and proposes strategies to enhance LPR for better robustness, interpretability, and transferability.","arXiv :2306 . 15638v1 [physics .chem-ph] 27 Jun 2023  \nRobustness of Local Predictions in Atomistic Machine Learning Models  \nSanggyu Chong,1 Federico Grasselli,1 Chiheb Ben Mahmoud,1 Joe D. Morrow,2 Volker L. Deringer,2 and Michele Ceriotti1  \n1) Laboratory of Computational Science and Modeling, Institute of Materials,  \n􀀓  \nEcole Polytechnique F􀀓ed􀀓erale de Lausanne, 1015 Lausanne, Switzerland  \n2) Department of Chemistry, Inorganic Chemistry Laboratory, University of Oxford, Oxford OX1 3QR, United Kingdom  \nMachine learning (ML) models for molecules and materials commonly rely on a decomposition of the global target quantity into local, atom-centered contributions. This approach is convenient from a computational perspective, enabling large-scale ML-driven simulations with a linear-scaling cost, and also allow for the identi􀀌cation and post-hoc interpretation of contributions from individual chemical environments and motifs to complicated macroscopic properties. However, even though there exist practical justi􀀌cations for these decompositions, only the global quantity is rigorously de􀀌ned, and thus it is unclear to what extent the atomistic terms predicted by the model can be trusted. Here, we introduce a quantitative metric, which we call the local prediction rigidity (LPR), that allows one to assess how robust the locally decomposed predictions of ML models are. We investigate the dependence of LPR on the aspects of model training, particularly the composition of training dataset, for a range of di􀀋erent problems from simple toy models to real chemical systems. We present strategies to systematically enhance the LPR, which can be used to improve the robustness, interpretability, and transferability of atomistic ML models.  \nI. INTRODUCTION  \nExtensive properties of matter, such as the total energy, arise from the collective interactions between atoms and can only be rigorously de􀀌ned as global quantities that depend on the entire molecule or condensed-phase structure. Nonetheless, the last decades have seen considerable e􀀋orts toward the construction of quantum-chemical methods that exploit the quantum-mechanical nearsightedness principle 1 to perform a local decomposition of the global quantities.2{7 These methods either undertake physically motivated local decomposition in the calculation of a global quantity,8{11 or perform such decomposition for the purpose of analysis. 12,13 Despite the fact that the local quantities are not physical observables, such a decomposition allows one to breakdown the macroscopic observable for a complex structure into contributions from much simpler components, typically individiual atoms. Consequently, such methods have led to drastic improvements in the time and cost scaling of quantum-mechanical calculations, and allowed researchers to gain an enhanced understanding of the physical and chemical nature of materials. 14{18 For example, symmetry-adapted perturbation theory (SAPT) is widely used to analyze non-covalent interactions between chemical species, 19 and projections from delocalized plane-wave basis sets into auxiliary atomic orbitals can be used to routinely study bonding and antibonding interactions in extended systems.20{22  \nThe idea of decomposing a global quantity into contributions associated with local environments has also become a cornerstone of atomistic machine learning (ML) .23{27 By \\learning\" the global quantities as a sum of local contributions, ML models can be trained to make predictions at the local level, which are then summed up to yield the global quantity prediction fora target system. Within the context of ML, this ap-  \nproach has two distinct advantages. First, it makes the ML models transferable and scalable, allowing them to be easily applied to systems of vastly di􀀋erent length scales (training on small cells, predicting for much larger ones),23,24 which underpins their pronounced success. Second, a local \\machine-learned\"decomposition of the global qu","cbCairKepQtPwHJB","https://ap.wps.com/l/cbCairKepQtPwHJB","pdf",7290788,1,13,"English","en",105,"# Introduction\n## Motivation: local decomposition in atomistic ML\n## Global vs local prediction reliability\n## Proposed metric: local prediction rigidity (LPR)","[{\"question\":\"What problem does the paper address in atomistic machine learning?\",\"answer\":\"It addresses the lack of rigorous guarantees for atom-centered local contributions, since only the global quantity is strictly defined in these decompositions.\"},{\"question\":\"What is local prediction rigidity (LPR)?\",\"answer\":\"LPR is a quantitative metric introduced to assess how robust the locally decomposed predictions of atomistic ML models are.\"},{\"question\":\"How do the authors improve the robustness of local predictions?\",\"answer\":\"They investigate how LPR depends on model training, especially training dataset composition, and propose strategies to systematically enhance LPR to improve robustness, interpretability, and transferability.\"}]","Robustness of Local Predictions in Atomistic Machine Learning Models - Paper | PDF",1785900189,33,{"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},"robustness-of-local-predictions-in-atomistic-machine-learning-models-paper","",{"@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/robustness-of-local-predictions-in-atomistic-machine-learning-models-paper/125606/",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},"What problem does the paper address in atomistic machine learning?","Question",{"text":75,"@type":76},"It addresses the lack of rigorous guarantees for atom-centered local contributions, since only the global quantity is strictly defined in these decompositions.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What is local prediction rigidity (LPR)?",{"text":80,"@type":76},"LPR is a quantitative metric introduced to assess how robust the locally decomposed predictions of atomistic ML models are.",{"name":82,"@type":73,"acceptedAnswer":83},"How do the authors improve the robustness of local predictions?",{"text":84,"@type":76},"They investigate how LPR depends on model training, especially training dataset composition, and propose strategies to systematically enhance LPR to improve robustness, interpretability, and transferability.","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"]