DocumentCode :
2952152
Title :
Combined Density Functional Theory and Ensembled Elman Network Correction Approach for Electronic Excitation Energies
Author :
Li, Hui ; Gao, Ting ; Lu, Yinghua ; Li, Hongzhi ; Su, Zhongmin
Author_Institution :
Sch. of Comput. Sci. & Inf. Technol., Northeast Normal Univ., Changchun, China
fYear :
2011
fDate :
30-31 July 2011
Firstpage :
1
Lastpage :
4
Abstract :
An ensemble of Elman networks (EEN) is formed by bagging to enhance the performance of the individual networks. The combined density functional theory (DFT) with EEN correction approach has been applied to evaluate the electronic excitation energies of organic molecules. The EEN approach improved DFT calculation results and reduced the RMS deviations from 0.48 to 0.23 eV for the training set. For the testing set, it is reduced from 0.41 to 0.22 eV. In general, the EEN approach leads to better results and shows the good generalization ability.
Keywords :
physics computing; recurrent neural nets; density functional theory; discrete Fourier transforms; electronic excitation energy; ensembled Elman network correction approach; organic molecule; Accuracy; Artificial neural networks; Bagging; Context; Discrete Fourier transforms; Testing; Training;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Control, Automation and Systems Engineering (CASE), 2011 International Conference on
Conference_Location :
Singapore
Print_ISBN :
978-1-4577-0859-6
Type :
conf
DOI :
10.1109/ICCASE.2011.5997564
Filename :
5997564
Link To Document :
بازگشت