Elm-based Time-variant Neural Networks with Incremental Number of Output Basis Functions

An Extreme Learning Machine (ELM) approach has already been applied to Time-Variant Neural Networks (TV-NN) with greatly reduced training time. However, several parameters need to be tuned in ELM-TV-NN, such as number of hidden neurons, number of basis functions. Interesting approaches have been proposed to automatically determine the number of hidden nodes in our previous work. In this paper, we explored a way to extend the Error Minimized Extreme Learning Machine (EM-ELM) algorithm along with one incremental based ELM method to the output basis functions case study. Simulation results show the effectiveness of the approach.