amrElm-package: Implementation of AMR-ELM (Affinity Matrix Regularized...

In rladeira/amrElm: Affinity Matrix Regularized Extreme Learning Machine

Description

This package implements a regularized version of ELM. Regularization is performed using a priori spacial information expressed by an affinity matrix. The use of this type of a priori information is similar to perform Tikhonov regularization. Furthermore, if a parameter free affinity matrix is used, like the cosine similarity matrix, regularization is performed without any need for parameter tunning. This version is applicable for binary classification problems.

Details

Package:	amrElm
Type:	Package
Version:	1.0
Date:	2014-12-03
License:	MIT

To create the model, the function to use is amrElmTrain. To predict values (or to test the model), the function to use is amrElmTest.

Author(s)

Leonardo José Silvestre

Maintainer: Leonardo José Silvestre <[e-mail]>

References

SILVESTRE, L. J. ; LEMOS, A. P. ; BRAGA, J. P. ; Braga, A. P. . Dataset Structure as Prior Information for Parameter-Free Regularization of Extreme Learning Machines In: Neurocomputing (accepted for publication).

SILVESTRE, L. J. ; LEMOS, A. P. ; BRAGA, J. P. ; Braga, A. P. . Parameter-free regularization in Extreme Learning Machines with affinity matrices. In: Proceedings of the European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, ESANN 2014, 23-25 April 2014, Bruges, Belgium. Louvain-la-Nueve, Belgique, 2014. p. 595-600.

rladeira/amrElm documentation built on May 27, 2019, 9:17 a.m.