Title :
Convergence properties of an online learning algorithm in neural network models of complex systems
Author :
Azarskov, V.N. ; Nikolaienko, S.A. ; Zhiteckii, L.S.
Author_Institution :
Aircraft Control Syst. Dept., Nat. Aviation Univ., Kiev, Ukraine
Abstract :
Asymptotic behavior of the online gradient algorithm with a constant step size employed for learning in neural network models of nonlinear systems having hidden layer are studied. The sufficient conditions guaranteeing the convergence of this algorithm in the random environment are established.
Keywords :
convergence; large-scale systems; learning (artificial intelligence); neural nets; asymptotic behavior; complex systems; constant step size; convergence properties; hidden layer; neural network models; nonlinear systems; online gradient algorithm; random environment; sufficient conditions; Artificial neural networks; Biological neural networks; Conferences; Convergence; Neurons; Unmanned aerial vehicles; convergence; gradient algorithm; learning; neural network; nonlinear model;
Conference_Titel :
Actual Problems of Unmanned Air Vehicles Developments Proceedings (APUAVD), 2013 IEEE 2nd International Conference
Conference_Location :
Kiev
Print_ISBN :
978-1-4799-3305-1
DOI :
10.1109/APUAVD.2013.6705293
