As a guest user you are not logged in or recognized by your IP address. You have
access to the Front Matter, Abstracts, Author Index, Subject Index and the full
text of Open Access publications.
A random vector functional-link (RVFL) net (also called Extreme Learning Machine, ELM) is a fast learning algorithm using single hidden layer feedforward neural networks (SLFNs). However, RVFL/ELM usually needs large scale hidden layer nodes and requires many memory resources. This paper proposes a sparse single layer feedforward network (SSLFN) with better generalization performance for multi-classification. In SSLFNs, the number of hidden layer nodes is not randomly set, but is related to the classes. For a k-class problem, the number of hidden layer nodes is simply set to k. Then, the function at each node is optimized so as to improve its generalization performance. Experimental results on HCL2000, MNIST and USPS benchmark datasets show the classification performance of SSLFNs with 10 hidden layer nodes has outperformed that of RVFL/ELM with thousands of hidden layer nodes.
This website uses cookies
We use cookies to provide you with the best possible experience. They also allow us to analyze user behavior in order to constantly improve the website for you. Info about the privacy policy of IOS Press.
This website uses cookies
We use cookies to provide you with the best possible experience. They also allow us to analyze user behavior in order to constantly improve the website for you. Info about the privacy policy of IOS Press.