Title :
L2 approximation properties of recurrent neural networks
Author :
Ruiz, A. ; Owens, D.H. ; Townley, S.
Author_Institution :
Centre for Syst. & Control Eng., Univ. of Exeter, Exeter, UK
Abstract :
We introduce a class of recurrent neural network that learn and replicates a periodic L2 function. A gradient type algorithm is used to modify the parameters of the network so that it learns and replicates autonomously a periodic signal. An example illustrates the results.
Keywords :
approximation theory; learning (artificial intelligence); recurrent neural nets; L2 approximation properties; autonomous periodic signal learning; autonomous periodic signal replication; gradient type algorithm; network parameters; periodic L2 function; recurrent neural networks; Approximation methods; Biological neural networks; Heuristic algorithms; Limit-cycles; Minimization; Recurrent neural networks; Roads; Intelligent Control; Neural Networks; Nonlinear Dynamics; Periodic Systems; Recurrent Neural Networks;
Conference_Titel :
Control Conference (ECC), 1997 European
Conference_Location :
Brussels
Print_ISBN :
978-3-9524269-0-6
