^O L 3@

   3 Level Hierarchy Object Learning Demonstration
   -----------------------------------------------

 This demonstration consists of a series of 3 recognizers
 that are capable of ^Object@ learning.  Each recognizer
 is based on the stimulus --> response model with the
 response of one recognizer used to create the input of
 another recognizer.

 Each recognizer contains an internal 63-cell memory that
 is modified during learning.  The position of a cell,
 when one is triggered, determines the value of the
 output response.

 The ^Demo@ option provides an easy way to get started.
 The ^Learn@ and ^Recognize@ options allow more user
 control.  The ^Display@ allows the user to observe
 the content of the internal memorys at any time.


 * demonstrates multi-level ^Object Learning and
   ^Recognition@ algorithm.  The level #2 learns
   in terms of the output from level #1, and level
   #3 learns in terms of the output from level #2.

 * the data input data is internally generated as a
   stream of random values with fixed 2 fixed patterns
   superimposed at random intervals.

 * the stimulus format consists of 8 elements per
   frame.  Each element is a number in the range:
   [0..63].


                     ^ - end help -