Please use this identifier to cite or link to this item:
|Title:||A study of extended minimal resource allocation network (EMRAN) alogrithm for multi-category classification problems||Authors:||P. Karuppasami||Keywords:||DRNTU::Engineering::Electrical and electronic engineering::Computer hardware, software and systems||Issue Date:||2011||Abstract:||This dissertation presents a Neural Network (NN) for multi-category classification problems and also presents the use of sequential learning algorithm, Extended Minimal Resource Allocating Network (EMRAN), to approximate the functional relationship between feature vector and class label. EMRAN uses Radial Basis Function (RBF) as its basic component. It has got growing and pruning strategy to find optimal hidden neurons. Its learning is based on Extended Kalman Filter (EKF) algorithm. In order to reduce the computational complexity, it updates only the winner neuron parameters.||Description:||67 p.||URI:||http://hdl.handle.net/10356/46916||Rights:||Nanyang Technological University||Fulltext Permission:||restricted||Fulltext Availability:||With Fulltext|
|Appears in Collections:||EEE Theses|
Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.