The analysis of genetic diseases has classically been directed towards establishing direct links between cause, a genetic variation, and effect, the observable deviation of phenot...
Benjamin Georgi, M. Anne Spence, Pamela Flodman, A...
Abstract. Sigmoidal or radial transfer functions do not guarantee the best generalization nor fast learning of neural networks. Families of parameterized transfer functions provide...
In this effort, we propose a new image fusion technique, utilizing Empirical Mode Decomposition (EMD), for improved face recognition. EMD is a non-parametric datadriven analysis t...
Harishwaran Hariharan, Andreas Koschan, Besma R. A...
—The paper describes the transfer functions with monotonic step response. These functions required in design of wide-band pulse amplifiers were synthesized using the first semi-p...
The choice of transfer functions may strongly influence complexity and performance of neural networks used in classification and approximation tasks. A taxonomy of activation an...