FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3084211373> ?p ?o ?g. }

• W3084211373 endingPage "104105" @default.
• W3084211373 startingPage "104105" @default.
• W3084211373 abstract "It is well known in the field of machine learning that committee models improve accuracy, provide generalization error estimates, and enable active learning strategies. In this work, we adapt these concepts to interatomic potentials based on artificial neural networks. Instead of a single model, multiple models that share the same atomic environment descriptors yield an average that outperforms its individual members as well as a measure of the generalization error in the form of the committee disagreement. We not only use this disagreement to identify the most relevant configurations to build up the model's training set in an active learning procedure, but also monitor and bias it during simulations to control the generalization error. This facilitates the adaptive development of committee neural network potentials and their training sets, while keeping the number of ab initio calculations to a minimum. To illustrate the benefits of this methodology, we apply it to the development of a committee model for water in the condensed phase. Starting from a single reference ab initio simulation, we use active learning to expand into new state points and to describe the quantum nature of the nuclei. The final model, trained on 814 reference calculations, yields excellent results under a range of conditions, from liquid water at ambient and elevated temperatures and pressures to different phases of ice, and the air-water interface - all including nuclear quantum effects. This approach to committee models will enable the systematic development of robust machine learning models for a broad range of systems." @default.
• W3084211373 created "2020-09-14" @default.
• W3084211373 creator A5004066800 @default.
• W3084211373 creator A5026599330 @default.
• W3084211373 creator A5059769803 @default.
• W3084211373 date "2020-09-14" @default.
• W3084211373 modified "2023-10-10" @default.
• W3084211373 title "Committee neural network potentials control generalization errors and enable active learning" @default.
• W3084211373 cites W1810694562 @default.
• W3084211373 cites W1975997599 @default.
• W3084211373 cites W1981368803 @default.
• W3084211373 cites W2008423326 @default.
• W3084211373 cites W2014166972 @default.
• W3084211373 cites W2025444507 @default.
• W3084211373 cites W2034637379 @default.
• W3084211373 cites W2037672122 @default.
• W3084211373 cites W2047524879 @default.
• W3084211373 cites W2052632546 @default.
• W3084211373 cites W2057858097 @default.
• W3084211373 cites W2058370262 @default.
• W3084211373 cites W2068335058 @default.
• W3084211373 cites W2073040595 @default.
• W3084211373 cites W2080021732 @default.
• W3084211373 cites W2083415705 @default.
• W3084211373 cites W2092157292 @default.
• W3084211373 cites W2093464914 @default.
• W3084211373 cites W2104489082 @default.
• W3084211373 cites W2135293965 @default.
• W3084211373 cites W2146440428 @default.
• W3084211373 cites W2155806188 @default.
• W3084211373 cites W2197007850 @default.
• W3084211373 cites W2200589053 @default.
• W3084211373 cites W2314762912 @default.
• W3084211373 cites W2316524229 @default.
• W3084211373 cites W2326801504 @default.
• W3084211373 cites W2410722695 @default.
• W3084211373 cites W2470768373 @default.
• W3084211373 cites W2524276051 @default.
• W3084211373 cites W2527189750 @default.
• W3084211373 cites W2538073050 @default.
• W3084211373 cites W2542537453 @default.
• W3084211373 cites W2547447472 @default.
• W3084211373 cites W2585152223 @default.
• W3084211373 cites W2592884795 @default.
• W3084211373 cites W2600764488 @default.
• W3084211373 cites W26088913 @default.
• W3084211373 cites W2620687153 @default.
• W3084211373 cites W2746244909 @default.
• W3084211373 cites W2749006386 @default.
• W3084211373 cites W2760908003 @default.
• W3084211373 cites W2765459427 @default.
• W3084211373 cites W2769254256 @default.
• W3084211373 cites W2771888471 @default.
• W3084211373 cites W28412257 @default.
• W3084211373 cites W2884430236 @default.
• W3084211373 cites W2890149186 @default.
• W3084211373 cites W2899557352 @default.
• W3084211373 cites W2912934387 @default.
• W3084211373 cites W2923693308 @default.
• W3084211373 cites W2939169979 @default.
• W3084211373 cites W2964268718 @default.
• W3084211373 cites W2970401211 @default.
• W3084211373 cites W2971894235 @default.
• W3084211373 cites W3002142428 @default.
• W3084211373 cites W3004981459 @default.
• W3084211373 cites W3005892601 @default.
• W3084211373 cites W3015526228 @default.
• W3084211373 cites W3103082775 @default.
• W3084211373 cites W4239414618 @default.
• W3084211373 doi "https://doi.org/10.1063/5.0016004" @default.
• W3084211373 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/32933264" @default.
• W3084211373 hasPublicationYear "2020" @default.
• W3084211373 type Work @default.
• W3084211373 sameAs 3084211373 @default.
• W3084211373 citedByCount "56" @default.
• W3084211373 countsByYear W30842113732020 @default.
• W3084211373 countsByYear W30842113732021 @default.
• W3084211373 countsByYear W30842113732022 @default.
• W3084211373 countsByYear W30842113732023 @default.
• W3084211373 crossrefType "journal-article" @default.
• W3084211373 hasAuthorship W3084211373A5004066800 @default.
• W3084211373 hasAuthorship W3084211373A5026599330 @default.
• W3084211373 hasAuthorship W3084211373A5059769803 @default.
• W3084211373 hasBestOaLocation W30842113732 @default.
• W3084211373 hasConcept C119857082 @default.
• W3084211373 hasConcept C134306372 @default.
• W3084211373 hasConcept C154945302 @default.
• W3084211373 hasConcept C177148314 @default.
• W3084211373 hasConcept C177264268 @default.
• W3084211373 hasConcept C178790620 @default.
• W3084211373 hasConcept C185592680 @default.
• W3084211373 hasConcept C199360897 @default.
• W3084211373 hasConcept C202444582 @default.
• W3084211373 hasConcept C2781442258 @default.
• W3084211373 hasConcept C33923547 @default.
• W3084211373 hasConcept C41008148 @default.
• W3084211373 hasConcept C50644808 @default.
• W3084211373 hasConcept C9652623 @default.

• next