Convergence properties of an online learning algorithm in neural network models of complex systems

Convergence properties of an online learning algorithm in neural network models of complex systems

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.

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Azarskov, V. N., Nikolaienko, S. A., Zhiteckii, L. S.
Format: Tagungsbericht
Sprache:eng
Schlagworte:
Artificial neural networks
Biological neural networks
Conferences
Convergence
gradient algorithm
learning
neural network
Neurons
nonlinear model
Unmanned aerial vehicles
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung: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.
DOI:10.1109/APUAVD.2013.6705293