Matches in SemOpenAlex for { <https://semopenalex.org/work/W4318613037> ?p ?o ?g. }
- W4318613037 endingPage "5396" @default.
- W4318613037 startingPage "5383" @default.
- W4318613037 abstract "Developing machine learning-based interatomic potentials from ab initio electronic structure methods remains a challenging task for computational chemistry and materials science. This work studies the capability of transfer learning, in particular discriminative fine-tuning, for efficiently generating chemically accurate interatomic neural network potentials on organic molecules from the MD17 and ANI data sets. We show that pre-training the network parameters on data obtained from density functional calculations considerably improves the sample efficiency of models trained on more accurate ab initio data. Additionally, we show that fine-tuning with energy labels alone can suffice to obtain accurate atomic forces and run large-scale atomistic simulations, provided a well-designed fine-tuning data set. We also investigate possible limitations of transfer learning, especially regarding the design and size of the pre-training and fine-tuning data sets. Finally, we provide GM-NN potentials pre-trained and fine-tuned on the ANI-1x and ANI-1ccx data sets, which can easily be fine-tuned on and applied to organic molecules." @default.
- W4318613037 created "2023-01-31" @default.
- W4318613037 creator A5022932307 @default.
- W4318613037 creator A5036807127 @default.
- W4318613037 creator A5047160329 @default.
- W4318613037 creator A5056979833 @default.
- W4318613037 date "2023-01-01" @default.
- W4318613037 modified "2023-10-10" @default.
- W4318613037 title "Transfer learning for chemically accurate interatomic neural network potentials" @default.
- W4318613037 cites W1582774210 @default.
- W4318613037 cites W1971044734 @default.
- W4318613037 cites W1985272737 @default.
- W4318613037 cites W2008489596 @default.
- W4318613037 cites W2012458244 @default.
- W4318613037 cites W2021362982 @default.
- W4318613037 cites W2025444507 @default.
- W4318613037 cites W2043578603 @default.
- W4318613037 cites W2091096906 @default.
- W4318613037 cites W2098614082 @default.
- W4318613037 cites W2527189750 @default.
- W4318613037 cites W2541404351 @default.
- W4318613037 cites W2585152223 @default.
- W4318613037 cites W2601081289 @default.
- W4318613037 cites W2749006386 @default.
- W4318613037 cites W2749580687 @default.
- W4318613037 cites W2785813126 @default.
- W4318613037 cites W2886916841 @default.
- W4318613037 cites W2892439819 @default.
- W4318613037 cites W2935741399 @default.
- W4318613037 cites W2954088480 @default.
- W4318613037 cites W2959623824 @default.
- W4318613037 cites W2962834855 @default.
- W4318613037 cites W2964268718 @default.
- W4318613037 cites W2990174420 @default.
- W4318613037 cites W3004981459 @default.
- W4318613037 cites W3010488723 @default.
- W4318613037 cites W3022256936 @default.
- W4318613037 cites W3031028823 @default.
- W4318613037 cites W3039722981 @default.
- W4318613037 cites W3043300003 @default.
- W4318613037 cites W3090555547 @default.
- W4318613037 cites W3098862241 @default.
- W4318613037 cites W3099276598 @default.
- W4318613037 cites W3102448310 @default.
- W4318613037 cites W3106310231 @default.
- W4318613037 cites W3133931590 @default.
- W4318613037 cites W3189164715 @default.
- W4318613037 cites W3189651386 @default.
- W4318613037 cites W3199869894 @default.
- W4318613037 cites W3200588771 @default.
- W4318613037 cites W3216896057 @default.
- W4318613037 cites W4213432497 @default.
- W4318613037 cites W4285293017 @default.
- W4318613037 cites W4307178223 @default.
- W4318613037 doi "https://doi.org/10.1039/d2cp05793j" @default.
- W4318613037 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/36748821" @default.
- W4318613037 hasPublicationYear "2023" @default.
- W4318613037 type Work @default.
- W4318613037 citedByCount "2" @default.
- W4318613037 countsByYear W43186130372023 @default.
- W4318613037 crossrefType "journal-article" @default.
- W4318613037 hasAuthorship W4318613037A5022932307 @default.
- W4318613037 hasAuthorship W4318613037A5036807127 @default.
- W4318613037 hasAuthorship W4318613037A5047160329 @default.
- W4318613037 hasAuthorship W4318613037A5056979833 @default.
- W4318613037 hasBestOaLocation W43186130372 @default.
- W4318613037 hasConcept C119857082 @default.
- W4318613037 hasConcept C121332964 @default.
- W4318613037 hasConcept C147597530 @default.
- W4318613037 hasConcept C150899416 @default.
- W4318613037 hasConcept C154945302 @default.
- W4318613037 hasConcept C177264268 @default.
- W4318613037 hasConcept C178790620 @default.
- W4318613037 hasConcept C185592680 @default.
- W4318613037 hasConcept C186060115 @default.
- W4318613037 hasConcept C18762648 @default.
- W4318613037 hasConcept C199360897 @default.
- W4318613037 hasConcept C2776372370 @default.
- W4318613037 hasConcept C2781442258 @default.
- W4318613037 hasConcept C41008148 @default.
- W4318613037 hasConcept C50644808 @default.
- W4318613037 hasConcept C58489278 @default.
- W4318613037 hasConcept C59593255 @default.
- W4318613037 hasConcept C62520636 @default.
- W4318613037 hasConcept C86803240 @default.
- W4318613037 hasConcept C97931131 @default.
- W4318613037 hasConceptScore W4318613037C119857082 @default.
- W4318613037 hasConceptScore W4318613037C121332964 @default.
- W4318613037 hasConceptScore W4318613037C147597530 @default.
- W4318613037 hasConceptScore W4318613037C150899416 @default.
- W4318613037 hasConceptScore W4318613037C154945302 @default.
- W4318613037 hasConceptScore W4318613037C177264268 @default.
- W4318613037 hasConceptScore W4318613037C178790620 @default.
- W4318613037 hasConceptScore W4318613037C185592680 @default.
- W4318613037 hasConceptScore W4318613037C186060115 @default.
- W4318613037 hasConceptScore W4318613037C18762648 @default.
- W4318613037 hasConceptScore W4318613037C199360897 @default.
- W4318613037 hasConceptScore W4318613037C2776372370 @default.