DocumentCode
1833257
Title
Reservoir optimization in recurrent neural networks using kronecker kernels
Author
Rad, Ali Ajdari ; Jalili, Mahdi ; Hasler, Martin
Author_Institution
Sch. of Comput. & Commun., Ecole Polytech. Fed. de Lausanne, Lausanne
fYear
2008
fDate
18-21 May 2008
Firstpage
868
Lastpage
871
Abstract
In this paper, using the mathematical properties of self-kronecker-production of small size random matrices, a simple but effective method is presented to optimize the reservoir of an echo state network given a certain task. The experimental results investigating the NARMA system show that few steps of the proposed optimization process can lead to a near optimum solution.
Keywords
learning (artificial intelligence); operating system kernels; optimisation; recurrent neural nets; NARMA system; kronecker kernels; recurrent neural networks; reservoir optimization; small size random matrices; Computer networks; Eigenvalues and eigenfunctions; Kernel; Optimization methods; Output feedback; Recurrent neural networks; Reservoirs; Sparse matrices; System testing; Transfer functions;
fLanguage
English
Publisher
ieee
Conference_Titel
Circuits and Systems, 2008. ISCAS 2008. IEEE International Symposium on
Conference_Location
Seattle, WA
Print_ISBN
978-1-4244-1683-7
Electronic_ISBN
978-1-4244-1684-4
Type
conf
DOI
10.1109/ISCAS.2008.4541556
Filename
4541556
Link To Document