TM ThinksProÔ

Neural Networks for Windows

Go to Home Page
Neural Network Products
Imaging Digital Photography Products

 

The Professional Neural Network Development Environment

 

    • Intuitive, interactive and highly responsive user interface
    • Built-in spreadsheets for convenient editing of data. Sensitivity Analysis and Input Disabling speed the development of high performance neural networks.
    • Dynamic graphing and Visualization Tools to view inputs, states, weights and outputs continuously illustrate training progress
    • Industry standard, documented, field-proven algorithms, make ThinksPro the perfect tool for students and professionals.
    • The first to offer reinforcement learning for complex control and financial applications. User defined grading eliminates the constraint of the fixed training set.
    • Ideal for forecasting, prediction, financial analysis, classification, function approximation, decision making, sensor interpretation, and control applications

 


¨ ThinksProTM provides maximum flexibility in seven categories
 
 

    • Preprocessing Functions
    • Interconnection Architectures
    • Processing Element Summation Functions
    • Processing Element Transfer Functions
    • Learning Algorithms
    • Learning Error Criteria
    • Visualization Tools

 

¨ Preprocessing Functions

          Thinks provides special input layer preprocessing to simplify working with all forms of data.

    • Min/Max and Mean/StdDev Input Preprocessing
    • Sum Inputs to 1 Normalization
    • Sum of Squares to 1 Normalization
    • Time Series Window Preprocessor
    • Input Disabling and Input Sensitivity Analysis

 

¨ Processing Element Summation Functions

          The function can be chosen independently on each layer of the network.
 

    • Dot Product
    • Quadratic Sum
    • L1 Distance
    • L2 Distance
    • Radial Basis Function
    • Sigma-Pi
    • GRNN Sum

 

¨ Processing Element Transfer Functions

          ThinksPro offers greater efficiency and training with 14 transfer functions.
 

    • Sigmoid
    • Bipolar Sigmoid
    • Arctan
    • Bipolar Arctan
    • Sin
    • Bipolar Sin
    • Threshold Linear
    • Bipolar Threshold Linear
    • Threshold
    • Bipolar Threshold
    • Linear
    • Gaussian
    • Cauchy
    • Winner Take All

 

¨ Interconnection Architectures

          Processing elements can be connected in networks up to five layers.
 

    • Multilayer Normal Feed Forward
    • Multilayer Full Feed Forward
    • Cascade
    • Total Recurrent
    • Jordan Recurrent
    • Elman Recurrent
    • Prior Recurrent
    • Cascade Recurrent

 

¨ Learning Algorithms

          Thinks supports fifteen industry standard algorithms:
 

    • Back Propagation
    • Quick Propagation
    • Jacob's Enhanced Back Propagation
    • Cascade Correlation Learning
    • Conjugate Gradients
    • Levenberg-Marquardt
    • Recurrent Back Propagation
    • Kohonen Winner Take All
    • Kohonen Learning Vector Quantization
    • Solis & Wets Simulated Annealing (graded)
    • Simplex Simulated Annealing (graded)
    • Powell’s Method Without Derivatives (graded)
    • Probabilistic Neural Network
    • General Regression Neural Network
    • Temporal Difference Learning

 

¨ Learning Error Criteria

           Control the network’s sensitivity to outliers in the training data with five error function options.
 

    • Mean Squared Error
    • Mean Absolute Error
    • Mean Fourth Power Error
    • Hyperbolic Square Error
    • Bipolar Hyperbolic Square Error
    • Classification Error
    • User Defined Grading Function

 

¨ Visualization Tools

        Examine the network inputs, weights and states during training with time series, Hinton and color diagrams.

 


ThinksPro contains the most sophisticated algorithms and capabilities available. ThinksPro provides multiple views into how neural networks learn and work. ThinksPro is the first to take you into the world of reinforcement learning with the EvalNet User Defined Grading Function and four graded learning algorithms, including the Temporal Difference Algorithm. ThinksPro comes standard with the following:

¨ Windows Dynamic Link Library (DLL)

Allows the professional user to build custom Windows, Visual Basic and Excel 5.0 applications.
 

¨ Sensitivity Analysis of Input Variables

Sensitivity Analysis facilitates the determination of the relative importance of the network inputs. Input disabling allows you to turn off unimportant inputs without reloading or rebuilding the training data.
 

¨ EvalNet User Defined Grading (DLL)

This allows the user to interactively test the network during the learning process. This allows ThinksPro to train networks for complex control applications or to maximize return on investment or any cost function in financial applications. This feature brings the full power of the ThinksPro visualization tools to your custom application.
 


To Order the Software…

Home

ThinksPro, EvalNet are trademarks of Logical Designs Consulting, Inc.
Windows, Visual Basic and Excel are trademarks of Microsoft Corporation.

Logical Designs Consulting, Inc. 2015 Olite Court, La Jolla, CA 92037 (858) 459-6236
 
 

Copyright 2002, Logical Designs Consulting, Inc.  All Rights Reserved