Interactive Newton's Method
This is a simple tool that demonstrates the overall concept of
optimization. The fundamental idea is: If a process can be defined parametrically, as a collection of inputs X, that lead to an output, f(X); then, a minimum (or maximum) of the output can be found by examining the gradients (changes of the inputs wrt the outputs) of the process at any set of inputs that is sufficiently close to the desired min/max. By applying Newton's Method, an iteration of better guesses can be generated, each in the beneficial direction toward that min/max. For example, consider the simple, one-dimensional polynomial of the form: |
A*(x+D)^3 + B*(x+D)^2 + C*(x+D) + Z |
Instructions:
|