Dynamic System Identification Using Recurrent Neural Networks

Pham, Duc Truong; Liu, Xing

doi:10.1007/978-1-4471-3244-8_3

Cited by 26 publications

(30 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Next, the activation of each hidden unit is copied into a corresponding context unit on a one-for-one basis with fixed weights of 1, and then the next time step is performed. This is equivalent to a recurrent connection from every hidden unit to itself and is more restrictive than the arbitrary recurrent connections allowed by Minsky's claim [31].…”

Section: Elman Networkmentioning

confidence: 99%

Ensemble ANNs-PSO-GA Approach for Day-ahead Stock E-exchange Prices Forecasting

Xiao¹,

Xiao²,

Lu³

et al. 2013

IJCIS

View full text Add to dashboard Cite

Stock e-exchange prices forecasting is an important financial problem that is receiving increasing attention. This study proposes a novel three-stage nonlinear ensemble model. In the proposed model, three different types of neural-network based models, i.e. Elman network, generalized regression neural network (GRNN) and wavelet neural network (WNN) are constructed by three non-overlapping training sets and are further optimized by improved particle swarm optimization (IPSO). Finally, a neural-network-based nonlinear meta-model is generated by learning three neural-network based models through support vector machines (SVM) neural network. The superiority of the proposed approach lies in its flexibility to account for potentially complex nonlinear relationships. Three daily stock indices time series are used for validating the forecasting model. Empirical results suggest the ensemble ANNs-PSO-GA approach can significantly improve the prediction performance over other individual models and linear combination models listed in this study.

show abstract

Section: Elman Networkmentioning

confidence: 99%

Ensemble ANNs-PSO-GA Approach for Day-ahead Stock E-exchange Prices Forecasting

Xiao¹,

Xiao²,

Lu³

et al. 2013

IJCIS

View full text Add to dashboard Cite

show abstract

“…Following by the literatures [17,18] and [20], we choose . Therefore, Equation (9) can be modified as…”

Section: A Description Of Hdnn Modelmentioning

confidence: 99%

“…HDNN can be easily realized by a Hopfield circuit and has the property of decreasing in energy by finite number of node-updating steps. In HDNN, the analysis of fundamental properties, stability, convergence and equilibrium for discrete and continuous systems were proposed in [17,18].…”

Section: Introductionmentioning

confidence: 99%

Intelligent switching adaptive control for uncertain nonlinear dynamical systems

Lee

Chiang

et al. 2015

Applied Soft Computing

View full text Add to dashboard Cite

“…All algebraic (feedforward) NNs, FNs, and WNs suffer from some drawbacks. In non-linear system modeling, a taped-delay lines approach is required, resulting in the number of rules increasing exponentially, the number of parameters in the rules getting large (this is called as ''the curse of dimensionality''), a long computational time, easily being affected by external noise, and difficulty in obtaining an independent system simulator [32,45,52,54]. The major drawbacks in these architectures are the curse of dimensionality, such as the requirement of too many parameters in NNs, the use of large rule bases in FL, the large number of wavelets, and the long training times, etc.…”

Section: Introductionmentioning

confidence: 99%

On three intelligent systems: dynamic neural, fuzzy, and wavelet networks for training trajectory

Becerikli

2004

Neural Comput & Applic

View full text Add to dashboard Cite

Intelligent systems cover a wide range of technologies related to hard sciences, such as modeling and control theory, and soft sciences, such as the artificial intelligence (AI). Intelligent systems, including neural networks (NNs), fuzzy logic (FL), and wavelet techniques, utilize the concepts of biological systems and human cognitive capabilities. These three systems have been recognized as a robust and attractive alternative to the some of the classical modeling and control methods. The application of classical NNs, FL, and wavelet technology to dynamic system modeling and control has been constrained by the non-dynamic nature of their popular architectures. The major drawbacks of these architectures are the curse of dimensionality, such as the requirement of too many parameters in NNs, the use of large rule bases in FL, the large number of wavelets, and the long training times, etc. These problems can be overcome with dynamic network structures, referred to as dynamic neural networks (DNNs), dynamic fuzzy networks (DFNs), and dynamic wavelet networks (DWNs), which have unconstrained connectivity and dynamic neural, fuzzy, and wavelet processing units, called ''neurons'', ''feurons'', and ''wavelons'', respectively. The structure of dynamic networks are based on Hopfield networks. Here, we present a comparative study of DNNs, DFNs, and DWNs for non-linear dynamical system modeling. All three dynamic networks have a lag dynamic, an activation function, and interconnection weights. The network weights are adjusted using fast training (optimization) algorithms (quasiNewton methods). Also, it has been shown that all dynamic networks can be effectively used in non-linear system modeling, and that DWNs result in the best capacity. But all networks have non-linearity properties in non-linear systems. In this study, all dynamic networks are considered as a non-linear optimization with dynamic equality constraints for non-linear system modeling. They encapsulate and generalize the target trajectories. The adjoint theory, whose computational complexity is significantly less than the direct method, has been used in the training of the networks. The updating of weights (identification of network parameters) is based on Broyden-Fletcher-Goldfarb-Shanno method. First, phase portrait examples are given. From this, it has been shown that they have oscillatory and chaotic properties. A dynamical system with discrete events is modeled using the above network structure. There is a localization property at discrete event instants for time and frequency in this example.

show abstract

Dynamic System Identification Using Recurrent Neural Networks

Cited by 26 publications

References 5 publications

Ensemble ANNs-PSO-GA Approach for Day-ahead Stock E-exchange Prices Forecasting

Ensemble ANNs-PSO-GA Approach for Day-ahead Stock E-exchange Prices Forecasting

Intelligent switching adaptive control for uncertain nonlinear dynamical systems

On three intelligent systems: dynamic neural, fuzzy, and wavelet networks for training trajectory

Contact Info

Product

Resources

About