Neuralis: an artificial neural network package Sami Khuri Dept. of Math, & Computer Science San Jose State University One Washington Square San Jose, CA 95192-0103 Jason Williams Absitomen Software 21845 Hermosa Ave. Cupertino, CA. 95014 73141 .2034 @compuserve.com khuri@sjsumcs.sjsu. Abstract edu understanding of the concepts and open the door for further investigation of the networks and their capabilities. Neuralis is basically a highly interactive, CAI (Computer Aided Instruction), self-paced program. Due to the flexibility of the package, one can use it to experiment with well-advanced topological networks of several hidden layers. The advanced user can provide his/her own custom-made parameters and incorporate them in the package before applying Neuralis to the challenging problem at hand. In the next section we introduce Neuralis and give an account of its efficiency, and power. The educational benefits that can be derived from using Neuralis, are fully discussed in the next and followirw sections. The conclusion with some final remarks and futur~ directions end this work. This work describes a highly interactive, self-paced package, Neuralis, that can be used in Artificial Intelligence courses in general, and Neural Networks in particular, Students sometimes tend to resist theory, preferring hands-on experience. Neuralis can be
/lp/association-for-computing-machinery/neuralis-an-artificial-neural-network-package-1ofsjXm7hH