Extreme Learning Machine with Randomly Assigned RBF Kernels

A new learning algorithm called extreme learning machine (ELM) has recently been proposed for single-hidden layer feedforward neural networks (SLFNs) with additive neurons to easily achieve good generalization performance at extremely fast learning speed. ELM randomly chooses the input weights and analytically determines the output weights of SLFNs. It is proved in theory that ELM can be extended to single-hidden layer feedforward neural networks (SLFNs) with radial basis function (RBF) kernels RBF networks, which allows the centers and impact widths of RBF kernels to be randomly generated and the output weights to be simply analytically calculated instead of iteratively tuned. The kernel function of ELM can be any nonlinear bounded integrable function which is almost continuous anywhere. Interestingly, the experimental results show that the ELM algorithm for RBF networks can complete learning at extremely fast speed and produce generalization performance very close to that of SVM in some benchmarking function approximation and classification problems. Index terms Radial basis function network, feedforward neural networks, real time learning, extreme learning machine, ELM, arbitrary kernels. The preliminary idea of the ELM algorithm for RBF networks has been shown in the Proceedings of the Eighth International Conference on Control, Automation, Robotics and Vision (ICARCV 2004), Dec 6-9, Kunming, China. Guang-Bin Huang and Chee-Kheong Siew Extreme Learning Machine with Randomly Assigned RBF Kernels