Accurate validation of GCV-based regularization parameter for extreme learning machine
Shraddha M. Naik, Ravi Prasad K. Jagannath · 2017
Extreme Learning Machine (ELM) is a neural network architecture with Single Layer Feed-forward Neural Network (SLFN). For meaningful results, the structure of ELM has to be optimized through the inclusion of regularization and the ℓ2- norm based regularization is mostly used. ℓ2-norm based regularization achieves better performance than the traditional ELM. The estimate of the regularization parameter is mainly through empirical methods or it is heuristically selected through prior experience. When such a choice is not possible, the Generalized Cross-Validation (GCV) method is one of a most popular choice for obtaining optimal regularization parameter. In this work, the Receiver Operating Characteristics (ROC) analysis is used to validate the regularization parameter obtained through GCV based method by evaluating the area under the ROC curve.