- You´ll need Java installed to view this page appropriately -
A global maximum is a point x* for which f(x*) ³ f(x) for all x, in other words, the largest overall value of a set, function, etc., over its entire range. Similarly, a global minimum is a point x* for which f(x*) £ f(x) for all x (the smallest overall value of a set, function, etc., over its entire range).[1] Any global maximum (minimum) is also a local maximum (minimum); however, a local maximum or minimum need not also be a global maximum or minimum. |
It is impossible to construct an algorithm that will find a global minimum/maximun for an arbitrary function [2]. Below, you will see a demonstration applet that use meta-heuristic that try to do this for continued functions.
My framework for metaheuristic algorithms has 5 until now: (update 28/02/2006)
| - Genetic Algortihm (for an introduction about genetic algorithms try this page) |
| - Simulated Annealing (SA) |
| - Hill-Climbing |
| - Iterated Local Search (ILS) |
| - Greedy Randomize Adaptative Procedure (GRASP) |
Launch it HERE (in your desktop with Java Web-Start) or click below to open the applet in a new page.
Send your comments about this applet via email: mrgatto@gmail.com
Create by Marcelo gaTTo
This applet is open-source, you can get it here.