This article demonstrates a new conceptor network based classifier in classifying images. Mathematical descriptions and analysis are presented. Various tests are experimented using three benchmark datasets: MNIST, CIFAR-10 and CIFAR-100. The experiments displayed that conceptor network can offer superior results and flexible configurations than conventional classifiers such as Softmax Regression and Support Vector Machine (SVM).
@article{arxiv.1506.00815,
title = {Classify Images with Conceptor Network},
author = {Yuhuang Hu and M. S. Ishwarya and Chu Kiong Loo},
journal= {arXiv preprint arXiv:1506.00815},
year = {2015}
}
Comments
This paper has been withdrawn by the author due to a crucial sign error in experiments