The Extreme Learning Machine (ELM) is a learning algorithm that first appeared in Guang-Bin Huang, Qin-Yu Zhu, Chee-Kheong Siew. "Extreme learning machine: Theory and applications", Neurocomputing (70), pp. 489 - 501, 2006 for single-hidden layer feedforward neural networks, which randomly chooses hidden nodes and analytically determines the output weights of the neural network. It has been reported to show good generalization performance at extremely fast learning speeds. Here, we implement the training algorithm as presented in the above paper, and also try it on one of the referenced datasets.
The full code and all outputs can be found in the ELM.pdf file