Recurrent Networks and NARMA Modeling
نویسندگان
چکیده
Douglas R. Martin B-317 Dept. of Statistics University of Washington Seattle, Washington 98195 There exist large classes of time series, such as those with nonlinear moving average components, that are not well modeled by feedforward networks or linear models, but can be modeled by recurrent networks. We show that recurrent neural networks are a type of nonlinear autoregressive-moving average (N ARMA) model. Practical ability will be shown in the results of a competition sponsored by the Puget Sound Power and Light Company, where the recurrent networks gave the best performance on electric load forecasting.
منابع مشابه
Recurrent neural networks and robust time series prediction
We propose a robust learning algorithm and apply it to recurrent neural networks. This algorithm is based on filtering outliers from the data and then estimating parameters from the filtered data. The filtering removes outliers from both the target function and the inputs of the neural network. The filtering is soft in that some outliers are neither completely rejected nor accepted. To show the...
متن کاملAdaptive control of nonlinear multivariable systems using neural networks
This paper considers the problem of using approximate methods for realizing the neural controllers for nonlinear multivariable systems. In [1] the NARMA-L1 and NARMA-L2 models were introduced as approximations of he NARMA model used for the representation of a SISO nonlinear dynamical systems. The advantage obtained from using NARMA-L1 and NARMA-L2 models is that control input u(k) occurs linea...
متن کاملNeural Network Based Predictive, Narma-l2 and Neuro-fuzzy Control for a Cstr Process
In recent years, there has been an expansive growth in the study and implementation of neural networks over a spectrum of research domains. Neural network based Predictive control is recognized as an efficient methodology to address difficult control problems. The NARMA model is an exact representation of the input-output behaviour of finite dimensional non-linear discrete time dynamical system...
متن کاملAdaptive Nonlinear System Identification with Echo State Networks
Echo state networks (ESN) are a novel approach to recurrent neural network training. An ESN consists of a large, fixed, recurrent "reservoir" network, from which the desired output is obtained by training suitable output connection weights. Determination of optimal output weights becomes a linear, uniquely solvable task of MSE minimization. This article reviews the basic ideas and describes an ...
متن کاملδ-NARMA neural networks: a new approach to signal prediction
This article presents a new connectionist architecture for stochastic univariate signal prediction. After a review of related statistical and connectionist models pointing out their advantages and limitations, we introduce the-NARMA model as the simplest non-linear extension of ARMA models. These models then provide the units of a MLP-like neural network: the-NARMA neural network. The associate...
متن کامل