This is a demonstration of gradient descent, line search
(steepest descent) and conjugate gradient methods.

At matlab prompt type 'optdemo', i.e.:
>> optdemo

Begin 	- initializes the simulation.
Run	- continuous run of the simulator
Step	- iterates through step of the simulation
Stop	- stops the continuous run.

Popup Menu:
Select either 'Gradient Descent','Line Search' or
'Conjugate Gradient' for the opimization method.
The optimization method may be changed 'on the fly'
during the simulation.

Sliders:
With Gradient Descent you can change the Learning Rate
and the Momentum both before and during the simulation.

Reset Button:
Press the 'Reset to Defaults' button to reset the simulator
to the default values of the Hessian, Momentum and Learning
Rate.

Hessian:
Edit the Hessian to create your own error surface. Remember
the Hessian has to be symmetric and positive definite
(positive eigenvalues).

