Search

Your search keyword '"Dorffner, Georg"' showing total 10 results
Start Over Author "Dorffner, Georg" Topic recurrent neural network
10 results on '"Dorffner, Georg"'

2. Controlling oscillatory behaviour of a two neuron recurrent neural network using inputs

Author

Haschke, Robert, Steil, Jochen J., Ritter, Helge, Dorffner, Georg, Bischof, Horst, and Hornik, Kurt

Subjects
Hopf bifurcation, symbols.namesake, Recurrent neural network, Artificial neural network, Control theory, Cellular neural network, Oscillation (cell signaling), symbols, Saddle-node bifurcation, Sigmoid function, Bifurcation, Mathematics
Abstract

We derive analytical expressions of codim-1-bifurcations for a fully connected, additive two-neuron network with sigmoidal activations, where the two external inputs are regarded as bifurcation parameters. The obtained Neimark-Sacker bifurcation curve encloses a region in input space with stable oscillatory behaviour, in which it is possible to control the oscillation frequency by adjusting the inputs.

Published
2001

8. Adaptive Neural Control of Nonlinear Systems

Author

Baruch, Ieroham, Flores, Jose Martin, Thomas, Federico, Garrido, Ruben, Goos, Gerhard, editor, Hartmanis, Juris, editor, van Leeuwen, Jan, editor, Dorffner, Georg, editor, Bischof, Horst, editor, and Hornik, Kurt, editor

Published
2001
Full Text
View/download PDF

10. Exponential Transients in Continuous-Time Symmetric Hopfield Nets

Author

Jirí Síma, Pekka Orponen, Dorffner, Georg, Bischof, Horst, and Hornik, Kurt

Subjects
Lyapunov function, Hopfield nets, stability, neural networks, Exponential function, Hopfield network, symbols.namesake, Models of neural computation, Recurrent neural network, Convergence (routing), symbols, Applied mathematics, Combinatorial optimization, dynaamiset systeemit, Algorithm, Mathematics, Network model
Abstract

We establish a fundamental result in the theory of continuous-time neural computation, by showing that so called continuous-time symmetric Hopfield nets, whose asymptotic convergence is always guaranteed by the existence of a Liapunov function may, in the worst case, possess a transient period that is exponential in the network size. The result stands in contrast to e.g. the use of such network models in combinatorial optimization applications. peerReviewed

Published
2001

Catalog

Books, media, physical & digital resources
See catalog results