Matches in SemOpenAlex for { <https://semopenalex.org/work/W3133762917> ?p ?o ?g. }
- W3133762917 abstract "Dynamics of flexible molecules are often determined by an interplay between local chemical bond fluctuations and conformational changes driven by long-range electrostatics and van der Waals interactions. This interplay between interactions yields complex potential-energy surfaces (PES) with multiple minima and transition paths between them. In this work, we assess the performance of state-of-the-art Machine Learning (ML) models, namely sGDML, SchNet, GAP/SOAP, and BPNN for reproducing such PES, while using limited amounts of reference data. As a benchmark, we use the cis to trans thermal relaxation in an azobenzene molecule, where at least three different transition mechanisms should be considered. Although GAP/SOAP, SchNet, and sGDML models can globally achieve chemical accuracy of 1 kcal mol-1 with fewer than 1000 training points, predictions greatly depend on the ML method used as well as the local region of the PES being sampled. Within a given ML method, large differences can be found between predictions of close-to-equilibrium and transition regions, as well as for different transition mechanisms. We identify key challenges that the ML models face in learning long-range interactions and the intrinsic limitations of commonly used atom-based descriptors. All in all, our results suggest switching from learning the entire PES within a single model to using multiple local models with optimized descriptors, training sets, and architectures for different parts of complex PES." @default.
- W3133762917 created "2021-03-15" @default.
- W3133762917 creator A5002697621 @default.
- W3133762917 creator A5026929463 @default.
- W3133762917 creator A5069777955 @default.
- W3133762917 creator A5087321869 @default.
- W3133762917 date "2021-03-03" @default.
- W3133762917 modified "2023-10-07" @default.
- W3133762917 title "Challenges for machine learning force fields in reproducing potential energy surfaces of flexible molecules" @default.
- W3133762917 cites W1531674615 @default.
- W3133762917 cites W1843396089 @default.
- W3133762917 cites W1973897761 @default.
- W3133762917 cites W1975755749 @default.
- W3133762917 cites W1975997599 @default.
- W3133762917 cites W1978183953 @default.
- W3133762917 cites W1981368803 @default.
- W3133762917 cites W1983132551 @default.
- W3133762917 cites W1988529738 @default.
- W3133762917 cites W1990531184 @default.
- W3133762917 cites W1998260904 @default.
- W3133762917 cites W2004677626 @default.
- W3133762917 cites W2016612689 @default.
- W3133762917 cites W2016893951 @default.
- W3133762917 cites W2023109668 @default.
- W3133762917 cites W2025444507 @default.
- W3133762917 cites W2028130802 @default.
- W3133762917 cites W2029413789 @default.
- W3133762917 cites W2031430816 @default.
- W3133762917 cites W2042315636 @default.
- W3133762917 cites W2048082410 @default.
- W3133762917 cites W2050796439 @default.
- W3133762917 cites W2057858097 @default.
- W3133762917 cites W2058370262 @default.
- W3133762917 cites W2059885388 @default.
- W3133762917 cites W2066130248 @default.
- W3133762917 cites W2072914432 @default.
- W3133762917 cites W2083415705 @default.
- W3133762917 cites W2084856161 @default.
- W3133762917 cites W2092144432 @default.
- W3133762917 cites W2104489082 @default.
- W3133762917 cites W2116906170 @default.
- W3133762917 cites W2148424525 @default.
- W3133762917 cites W2225949634 @default.
- W3133762917 cites W2337496963 @default.
- W3133762917 cites W2511310272 @default.
- W3133762917 cites W2547447472 @default.
- W3133762917 cites W2566573083 @default.
- W3133762917 cites W2585152223 @default.
- W3133762917 cites W2620687153 @default.
- W3133762917 cites W2650911154 @default.
- W3133762917 cites W2725390203 @default.
- W3133762917 cites W2730802076 @default.
- W3133762917 cites W2742127985 @default.
- W3133762917 cites W2749006386 @default.
- W3133762917 cites W2753962198 @default.
- W3133762917 cites W2757878424 @default.
- W3133762917 cites W2768213699 @default.
- W3133762917 cites W2774480125 @default.
- W3133762917 cites W2778051509 @default.
- W3133762917 cites W2782772320 @default.
- W3133762917 cites W2798718650 @default.
- W3133762917 cites W2800440295 @default.
- W3133762917 cites W2808939444 @default.
- W3133762917 cites W2886120629 @default.
- W3133762917 cites W2891365537 @default.
- W3133762917 cites W2904141086 @default.
- W3133762917 cites W2907807628 @default.
- W3133762917 cites W2934126989 @default.
- W3133762917 cites W2952916873 @default.
- W3133762917 cites W2954088480 @default.
- W3133762917 cites W2964268718 @default.
- W3133762917 cites W2972246420 @default.
- W3133762917 cites W2976213714 @default.
- W3133762917 cites W2991533556 @default.
- W3133762917 cites W3003486042 @default.
- W3133762917 cites W3024290514 @default.
- W3133762917 cites W3034519476 @default.
- W3133762917 cites W3092617600 @default.
- W3133762917 cites W3098341684 @default.
- W3133762917 cites W3099005864 @default.
- W3133762917 cites W3099423575 @default.
- W3133762917 cites W3099813870 @default.
- W3133762917 cites W3099950071 @default.
- W3133762917 cites W3101744125 @default.
- W3133762917 cites W3102448310 @default.
- W3133762917 cites W3103144868 @default.
- W3133762917 cites W3103502300 @default.
- W3133762917 cites W3103956262 @default.
- W3133762917 cites W3106310231 @default.
- W3133762917 cites W3125814979 @default.
- W3133762917 doi "https://doi.org/10.1063/5.0038516" @default.
- W3133762917 hasPublicationYear "2021" @default.
- W3133762917 type Work @default.
- W3133762917 sameAs 3133762917 @default.
- W3133762917 citedByCount "21" @default.
- W3133762917 countsByYear W31337629172021 @default.
- W3133762917 countsByYear W31337629172022 @default.
- W3133762917 countsByYear W31337629172023 @default.
- W3133762917 crossrefType "journal-article" @default.
- W3133762917 hasAuthorship W3133762917A5002697621 @default.