An approach to reservoir computing design and training

• Published in: Expert systems with applications Vol. 40; no. 10; pp. 4172 - 4182
• Main Authors: Ferreira, Aida A., Ludermir, Teresa B., de Aquino, Ronaldo R.B.
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier Ltd 01.08.2013; Elsevier
• Subjects: Adjustment; Applied sciences; Artificial intelligence; Computation; Computer science; control theory; systems; Computer systems and distributed systems. User interface; Connectionism. Neural networks; Design engineering; Dynamical systems; Echo state networks; Evolutionary algorithm; Exact sciences and technology; Inference from stochastic processes; time series analysis; Mathematical models; Mathematics; Probability and statistics; Recurrent neural networks; Reservoir computing; Reservoirs; Sciences and techniques of general use; Software; Statistics; Training; Reservoir computing; Evolutionary algorithm; Echo state networks; Recurrent neural nets; Spectral radius; Perron value; Time series; Black box; Reservoir; Systems theory; Neural network; Dynamical system; Topology; Non linear system; Distributed computing; Modeling; Weighted graph; Search algorithm; Genetic algorithm; Supervised learning; Occupation time; Learning algorithm
• ISSN: 0957-4174; 1873-6793
• DOI: 10.1016/j.eswa.2013.01.029

► We introduce a method, called RCDESIGN (reservoir computing and design training). ► RCDESIGN optimizes reservoir parameters, topology and reservoir weights simultaneously, without the spectral radius rescaling. ► RCDESIGN performs well, since the real dynamics of the reservoir can only be found with the system in use. ► We verified that the proposed method is good for forecasting four distinct time series. Reservoir computing is a framework for computation like a recurrent neural network that allows for the black box modeling of dynamical systems. In contrast to other recurrent neural network approaches, reservoir computing does not train the input and internal weights of the network, only the readout is trained. However it is necessary to adjust parameters to create a “good” reservoir for a given application. In this study we introduce a method, called RCDESIGN (reservoir computing and design training). RCDESIGN combines an evolutionary algorithm with reservoir computing and simultaneously looks for the best values of parameters, topology and weight matrices without rescaling the reservoir matrix by the spectral radius. The idea of adjust the spectral radius within the unit circle in the complex plane comes from the linear system theory. However, this argument does not necessarily apply to nonlinear systems, which is the case of reservoir computing. The results obtained with the proposed method are compared with results obtained by a genetic algorithm search for global parameters generation of reservoir computing. Four time series were used to validate RCDESIGN.