FUNCTION: 
  This program is a simulation of the cart and pole dynamic system and
  a procedure for learning to balance the pole.  Both are described in
  Barto, Sutton, and Anderson, "Neuronlike Adaptive Elements That Can
  Solve Difficult Learning Control Problems," IEEE Trans. Syst., Man,
  Cybern., Vol. SMC-13, pp. 834--846, Sept.--Oct. 1983, and in Sutton, 
  "Temporal Aspects of Credit Assignment in Reinforcement Learning", PhD
  Dissertation, Department of Computer and Information Science, University
  of Massachusetts, Amherst, 1984.

INPUT:
   There is one run-time inputs to this program: the seed value for the
random number generator.  To modify the reinforcement learning parameters
or the cart-pole simulation parameters, modify pole.h

OUTPUT:
   One line of output is generated per trial: the number of steps for
which the pole was balanced.   


INSTALLATION:

1. Edit the file Makefile, and change the line that begins:
	INCLUDE =

   to the c header include path on your system.
   (If you don't know the include path on your system, you might be able
    to find out with this command:
    SETENV | grep -i include

    If that doesn't work, ask your local sys. admin.)

2. Type make.  The executable will be called "pole"

3. Run "pole".  You will be prompted for the seed value, and then shown
   (in text mode) the trials. 

   If you would like to modify the reinforcement learning parameters 
   (such as the learning rate, discount factor, and decay rate), or the
   simulation parameters, such as the length of the pole, you can do
   so by editing pole.h, then doing a make (step 2) again.


BUGS AND LIMITATIONS:

  There is not yet a GUI interface to the cart-pole simulation.  Anyone,
   anyone?
   
