Improved recursive least squares algorithm based on echo state neural network for nonlinear system identification

In order to model nonlinear systems with more accuracy, and to further exploit the potential capacities of recurrent neural networks, we propose a novel recursive least square (RLS) algorithm based on echo state network (ESN), and note it as RLSESN in this paper. ESN is a new paradigm for using recu...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Proceedings of the 30th Chinese Control Conference s. 1692 - 1695
Hlavní autoři: Song Qingsong, Zhao Xiangmo, Feng Zuren
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 01.07.2011
Témata:
Adaptation models
Filtering
Modeling
Neural Networks
Neurons
Recurrent neural networks
Recursive Least Squares Algorithm
Reservoirs
System Identification
Training
ISBN:9781457706776, 1457706776
ISSN:1934-1768
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:In order to model nonlinear systems with more accuracy, and to further exploit the potential capacities of recurrent neural networks, we propose a novel recursive least square (RLS) algorithm based on echo state network (ESN), and note it as RLSESN in this paper. ESN is a new paradigm for using recurrent neural networks (RNN) with a simpler training method. The proposed RLSESN consists of three main components: an ESN, a recursive least square (RLS) algorithm with adaptive forgetting factor and a change detection module. At first, the change detection module modifies the forgetting factor online according to ESN output errors. And then, the RLS algorithm regulates the ESN output connection weights. The simulation experiment results show that RLSESN can model nonlinear systems very well; the modeling performances are significantly better than those traditional ARMA model based filters.
ISBN:9781457706776
1457706776
ISSN:1934-1768