Matches in SemOpenAlex for { <https://semopenalex.org/work/W4321614295> ?p ?o ?g. }
- W4321614295 abstract "Abstract Existing molecular property prediction methods based on deep learning ignore the generalization ability of the nonlinear representation of molecular features and the reasonable assignment of weights of molecular features, making it difficult to further improve the accuracy of molecular property prediction. To solve the above problems, an end-to-end double-head transformer neural network (DHTNN) is proposed in this paper for high-precision molecular property prediction. For the data distribution characteristics of the molecular dataset, DHTNN specially designs a new activation function, beaf, which can greatly improve the generalization ability of the nonlinear representation of molecular features. A residual network is introduced in the molecular encoding part to solve the gradient explosion problem and ensure that the model can converge quickly. The transformer based on double-head attention is used to extract molecular intrinsic detail features, and the weights are reasonably assigned for predicting molecular properties with high accuracy. Our model, which was tested on the MoleculeNet [1] benchmark dataset, showed significant performance improvements over other state-of-the-art methods." @default.
- W4321614295 created "2023-02-24" @default.
- W4321614295 creator A5010709242 @default.
- W4321614295 creator A5014254672 @default.
- W4321614295 creator A5014772779 @default.
- W4321614295 creator A5045067652 @default.
- W4321614295 creator A5048397321 @default.
- W4321614295 date "2023-02-23" @default.
- W4321614295 modified "2023-10-14" @default.
- W4321614295 title "Double-head transformer neural network for molecular property prediction" @default.
- W4321614295 cites W1481666845 @default.
- W4321614295 cites W1553697943 @default.
- W4321614295 cites W1975147762 @default.
- W4321614295 cites W1976526581 @default.
- W4321614295 cites W1993046136 @default.
- W4321614295 cites W2025580327 @default.
- W4321614295 cites W2028501442 @default.
- W4321614295 cites W2046589863 @default.
- W4321614295 cites W2065354989 @default.
- W4321614295 cites W2075613930 @default.
- W4321614295 cites W2085074871 @default.
- W4321614295 cites W2092285329 @default.
- W4321614295 cites W2096541451 @default.
- W4321614295 cites W2096560421 @default.
- W4321614295 cites W2114162221 @default.
- W4321614295 cites W2122825543 @default.
- W4321614295 cites W2145578524 @default.
- W4321614295 cites W2153564408 @default.
- W4321614295 cites W2276859037 @default.
- W4321614295 cites W2290847742 @default.
- W4321614295 cites W2461470610 @default.
- W4321614295 cites W2503169225 @default.
- W4321614295 cites W2527189750 @default.
- W4321614295 cites W2565684601 @default.
- W4321614295 cites W2594183968 @default.
- W4321614295 cites W2753962198 @default.
- W4321614295 cites W2769775068 @default.
- W4321614295 cites W2895420596 @default.
- W4321614295 cites W2901003004 @default.
- W4321614295 cites W2966357564 @default.
- W4321614295 cites W2969457089 @default.
- W4321614295 cites W2970852745 @default.
- W4321614295 cites W2972987306 @default.
- W4321614295 cites W3000478925 @default.
- W4321614295 cites W3012117454 @default.
- W4321614295 cites W3013625695 @default.
- W4321614295 cites W3018495986 @default.
- W4321614295 cites W3024033076 @default.
- W4321614295 cites W3038823641 @default.
- W4321614295 cites W3054891022 @default.
- W4321614295 cites W3113447514 @default.
- W4321614295 cites W3119774682 @default.
- W4321614295 cites W3129168016 @default.
- W4321614295 cites W3129577962 @default.
- W4321614295 cites W3163493952 @default.
- W4321614295 cites W3166272013 @default.
- W4321614295 cites W3198212015 @default.
- W4321614295 cites W3198599144 @default.
- W4321614295 cites W4210313485 @default.
- W4321614295 cites W4223955854 @default.
- W4321614295 cites W4226072475 @default.
- W4321614295 cites W4288275971 @default.
- W4321614295 cites W4289173507 @default.
- W4321614295 cites W4310709716 @default.
- W4321614295 doi "https://doi.org/10.1186/s13321-023-00700-4" @default.
- W4321614295 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/36823530" @default.
- W4321614295 hasPublicationYear "2023" @default.
- W4321614295 type Work @default.
- W4321614295 citedByCount "1" @default.
- W4321614295 countsByYear W43216142952023 @default.
- W4321614295 crossrefType "journal-article" @default.
- W4321614295 hasAuthorship W4321614295A5010709242 @default.
- W4321614295 hasAuthorship W4321614295A5014254672 @default.
- W4321614295 hasAuthorship W4321614295A5014772779 @default.
- W4321614295 hasAuthorship W4321614295A5045067652 @default.
- W4321614295 hasAuthorship W4321614295A5048397321 @default.
- W4321614295 hasBestOaLocation W43216142951 @default.
- W4321614295 hasConcept C111472728 @default.
- W4321614295 hasConcept C11413529 @default.
- W4321614295 hasConcept C119599485 @default.
- W4321614295 hasConcept C121332964 @default.
- W4321614295 hasConcept C124101348 @default.
- W4321614295 hasConcept C127413603 @default.
- W4321614295 hasConcept C13280743 @default.
- W4321614295 hasConcept C134306372 @default.
- W4321614295 hasConcept C138885662 @default.
- W4321614295 hasConcept C153180895 @default.
- W4321614295 hasConcept C154945302 @default.
- W4321614295 hasConcept C155512373 @default.
- W4321614295 hasConcept C158622935 @default.
- W4321614295 hasConcept C165801399 @default.
- W4321614295 hasConcept C177148314 @default.
- W4321614295 hasConcept C17744445 @default.
- W4321614295 hasConcept C185798385 @default.
- W4321614295 hasConcept C186060115 @default.
- W4321614295 hasConcept C189950617 @default.
- W4321614295 hasConcept C199539241 @default.
- W4321614295 hasConcept C205649164 @default.
- W4321614295 hasConcept C2776359362 @default.
- W4321614295 hasConcept C33923547 @default.