HNeT Discovery Package Version 1.3 by Lou Mendelsohn AND AMERICA LTD. 2140 Winston Park Drive, suite 202 Oakville, Ontario, Canada L6H 5V5 Phone: 905 829-0788 Fax: 905 829-0790 Product: Windows application designed to create, train and test neural networks using holographic neural technology. Equipment requirements: IBM compatible with an 80386 or higher processor. A 80387 math coprocessor recommended for 80386 or 80486 SX processors. 4 megabytes RAM required and minimum megabytes of hard drive space. VGA/XGA display adapter and monitor. Microsoft Windows 3.0 or higher; Microsoft-compatible mouse. Printer recommended but not required. Price: $295. The neural network system most frequently used for financial forecasting is the feedforward back-propagation network that is offered in most commercial neural network development tools such as BrainMaker and NeuralWorks. While this system has been successfully applied to a wide range of tasks, its training time on complex problems such as financial market forecasting can be considerable. Toward fulfilling the goal of reducing training time, a new development tool called HNeT Discovery Package sports a network paradigm referred to as holographic neural technology. Through the use of this system, HNeT finds near-optimal solutions in substantially less training time than back-propagation networks on similar applications. In addition, HNeT contains several other features of interest to neural network developers. |
FOR THOSE ORDERING ARTICLES SEPARATELY: *Note: $2.95-$5.95 Articles are in PDF format only. No hard copy of the article(s) will be delivered. During checkout, click the "Download Now" button to immediately receive your article(s) purchase. STOCKS & COMMODITIES magazine is delivered via mail.
|
