Title :
Convergence of batch gradient method with penalty and momentum for GNNM(1,1)
Author :
Zhengxue Li ; Yaowei Sun ; Mingsong Cheng ; Mingchen Yao
Author_Institution :
Sch. of Math. Sci., Dalian Univ. of Technol., Dalian, China
Abstract :
In this paper the penalty term and momentum term are used in batch gradient method which improve the learning method for one-dimensional gray neural network model (GNNM(1,1)). Moreover, the convergence of the modified batch gradient method for GNNM(1,1) is proved.
Keywords :
convergence; gradient methods; grey systems; learning (artificial intelligence); neural nets; batch gradient method convergence; learning method; momentum term; one-dimensional gray neural network model; penalty term; Accuracy; Convergence; Educational institutions; Gradient methods; Neural networks; Presses; Training; batch gradient method; convergence; momentum term; one-dimensional gray neural network; penalty term;
Conference_Titel :
Natural Computation (ICNC), 2013 Ninth International Conference on
Conference_Location :
Shenyang
DOI :
10.1109/ICNC.2013.6817934
