Backpropagation Simulation Assignment - CS 578

  Using  the  SNNS simulator, do the following experiments.   For
each  experiment  give a short (~paragraph)  discussion  of  your
findings   and  observations.   Remember.   This  is  a  powerful
network.   You're testing and experimenting with the real  thing.
Have  fun  with  it.  Most of the points are  for  your  work  in
numbers 2-4 below.

  Note:   For  all  assignments  in this  course  requiring  your
observations (which is the typical case) the most important thing
is  not  just mentioning what you observe occurring, but  I  want
your  explanation as to why it is occurring.  If you can't figure
it  out, then give your best try at explaining it.  This is where
learning can best occur, when trying to figure out why the models
do what they do.

1.  Using a simple network with one hidden layer do
   a)  Linearly Separable mappings
   b)  XOR function with different starting weights
             For the above:
             How long for learning?
             Will they always converge?
              What  is  the  effect of different LR and  momentum
   terms?

2.   Test  (both  training and generalization) some more  complex
   nets on a couple MLDB applications
   a)    Adjust  parameters such as initial weights,  lrate,  and
   momentum,  number  of hidden nodes, number of  hidden  layers.
   Note dynamics of TSS, weights, etc.
   b)  Is there some particularly difficult applications.  why?
   c)   Try  at  least  one application with  analog  inputs  and
   outputs

3.   Be creative.  Try some of the following and some of your own
   experiments:
   *   Come up with some of your own research results!

4.   Propose  your own real world application.  Gather a training
   and  test  set.   Train  it with a BP net.   Report  approach,
   results, accuracy, and observations.

Total pages: ~6-7 - This will be your most indepth simulation