EXTREME LEARNING MACHINES AND CLASSIFIER FUSION

Extreme learning machine (ELM) was proposed as a new efficient learning algorithm for single-hidden layer feed forward neural networks (SLFN) in recent years. It is featured by its much faster training speed, non-iterative training procedure and better generalized performance over traditional SLFN learning techniques. In this paper, we do a comparative study of how ELM stands against various Classification techniques, general performance and ELM performance in Ensemble Setting. We’ll compare the performance using various evaluation metrics.