Learning of a single-hidden layer feedforward neural network using an optimized extreme learning machine

• Published in: Neurocomputing (Amsterdam) Vol. 129; pp. 428 - 436
• Main Authors: Matias, Tiago, Souza, Francisco, Araújo, Rui, Antunes, Carlos Henggeler
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier B.V 10.04.2014; Elsevier
• Subjects: Algorithmics. Computability. Computer arithmetics; Applied sciences; Artificial intelligence; Computer science; control theory; systems; Computer simulation; Connectionism. Neural networks; Differential evolution; Evolution; Exact sciences and technology; Feedforward; Genetic algorithms; Learning; Neural networks; Optimization; Optimized extreme learning machine; Regularization; Simulated annealing; Single-hidden layer feedforward neural networks; Theoretical computing; Differential evolution; Single-hidden layer feedforward neural networks; Optimized extreme learning machine; Simulated annealing; Genetic algorithms; Evolutionary algorithm; Bias; Neural network; Optimization; Feedforward neural nets; Genetic algorithm; Least squares method; Hidden variable theory; Batch process; Feedforward; Learning algorithm; Regularization; Signal to noise ratio
• ISSN: 0925-2312; 1872-8286
• DOI: 10.1016/j.neucom.2013.09.016

This paper proposes a learning framework for single-hidden layer feedforward neural networks (SLFN) called optimized extreme learning machine (O-ELM). In O-ELM, the structure and the parameters of the SLFN are determined using an optimization method. The output weights, like in the batch ELM, are obtained by a least squares algorithm, but using Tikhonov's regularization in order to improve the SLFN performance in the presence of noisy data. The optimization method is used to the set of input variables, the hidden-layer configuration and bias, the input weights and Tikhonov's regularization factor. The proposed framework has been tested with three optimization methods (genetic algorithms, simulated annealing, and differential evolution) over 16 benchmark problems available in public repositories.