HOME BACKGROUND PRODUCTS APPLICATIONS SERVICES SUPPORT LINKS

Given numerical training data, nonlinear networks (including neural nets) learn how to map N-dimensional input vectors into M-dimensional output vectors. This makes nonlinear networks an ideal solution to a number of problems including pattern recognition, function approximation, data mining, and e-mail spam filtering. The diagram illustrates how our optimization system can use nonlinear networks to boost your application!

Click here for a PowerPoint Presentation Describing our work

User-provided Data Multilayer Perceptron Piecewise Linear Network Functional Link Network Nearest Neighbor Classifier Support Vector Machine Radial Basis Function Network Self Organizing Map Feature Selection Size Train Prune and Validate Final Network

Hover over diagram to view descriptions!


Copyright © Neural Decision Lab LLC