An incremental adaptive neural network model for online noisy data regression and its application to compartment fire studies


This paper presents a probabilistic-entropy-based neural network (PENN) model for tackling online data regression problems. The network learns online with an incremental growth network structure and performs regression in a noisy environment. The training samples presented to the model are clustered into hyperellipsoidal Gaussian kernels in the joint space… (More)
DOI: 10.1016/j.asoc.2010.01.002


12 Figures and Tables

Slides referencing similar topics