Back to DGP Applet page | Buttons | Windows | GP Settings |
The DGP applet allows you to control the genetic program and inspect the results.
- Start Button : Click on it to launch the genetic program.
- Pause/Resume Button : Click on it to hold and resume the genetic program.
- Stop Button : Click on it to abort the program.
- Settings Button : Use this to examine and change the parameters for the program.
- Continue running the applet when you leave the page (Yes/No) ? : Normally, the browser will suspend the applet thread when you leave the applet page to browse another page and resume execution when you return to the applet page. This option allows you to specify whether you want to run the applet in the background or suspend the applet when you leave the applet page. The default is "YES" which runs the applet in the background when you leave the page.
- Show the trail generated by the best program in the generation ? : This option allows you to specify whether you want to see an animated presentation showing the artificial ant turning left or right and moving along the trail generated by the best program in a generation. You can choose to see the presentation either at the end of the last generation of a run or at the end of every generation or you may choose not to see such a presentation at all. The default is to present the animated trail at the end of a run.
The 'Santa Fe Trail' window :
- The yellow and orange squares on the 32 x 32 toroidal grid show the Santa Fe trail to be followed by the artificial ant. The orange squares represent the 89 food units to be eaten by the ant.
- After launching the genetic program with the Start button, red lines appear to show the trail produced by the ant navigation program that is currently being evaluated.
- After each generation of the genetic program, black lines appear to show the trail generated by the ant navigation program with the highest fitness level (ie. the program that controls the ant to eat the most number of food units) in the previous generation.
- Round dots along a trail show the food eaten by the ant when it follows the trail produced by an ant navigation program.
- An animated presentation of the 'fittest' ant navigation program in the previous generation navigating the ant along the trail can be displayed either at the end of a run, after every generation or not displayed at all, depending on your choice.
The Fitnesses Window : When a complete generation has been evaluated, the Fitnesses window displays a histogram of the fitness values of all individuals.
The Status Window : The Status window indicates what is currently happening. Besides this, the fitnesses (hit counts/number of food pieces eaten) of the individuals (programs) sent to and received from the server are displayed when the applet sends and receives individuals from the server.
The Results Window : The Results window displays a symbolic representation of the best ant navigation program found so far, along with some other data.
You may inspect or change some parameters of the genetic
program in the Settings dialog. Call it up with the Settings
button.
Note that you cannot change the settings while the genetic
program is running. Therefore, the Settings button is enabled
only when the program is stopped.
Brief explanation of settings | |
---|---|
Population size | The number of individuals (i.e. programs) in the population. Each generation has the same number of individuals. |
Max no. of generations in a run | The maximum number of generations to be performed in a run. |
Number of generations between migrations | At an interval of a number of generations, the applet will connect to the server to do a migration whereby it exports a specified number of individuals to and imports a specified number of individuals from the server. This parameter defines this interval, in terms of the number of generations. |
Number of individuals to export | The number of individuals to export to the server during a migration. |
Number of individuals to import | The number of individuals to import from the server during a migration. |
Max. depth for new individuals | The maximum depth of program trees for new individuals. |
Crossover fraction Reproduction fraction Mutation fraction |
When producing a new
generation, new individuals are generated from the old
population by three methods:
The ratio of the fraction numbers determines the ratio of the methods chosen. |
Max. depth for individuals after crossover | The maximum tree depth for programs produced by crossover. If an offspring exceeds this limit, one of the parents is chosen instead. |
Max. depth for new subtrees in mutants | The maximum depth of subtrees that are spliced in at mutation. |
Method of generation | Determines the shape of the
program trees in the initial population:
|
Method of selection | Determines the method how
parents are selected when breeding a new generation:
|
Elitism | Specifies whether elitism is to be applied to the genetic program. With elitism, the best individual in the population of the current generation is always copied to the new population of the next generation to ensure that the best individual of a run is not lost through the generations. |
Fitness cases | For the Artificial Ant following the Santa Fe trail problem, there is only one fitness case. It is the Santa Fe trail which consists of 144 squares with 21 turns and 89 food units distributed non-uniformly along it. |
Function set | The list shows all the
functions available. Functions are the internal nodes of
the program trees. The algorithm uses only those functions that you have selected in the list. You must select at least one item. |
Terminal set | The list shows all the
terminal types available. Terminals are the leaves of the
program trees. The algorithm uses only those terminals that you have selected in the list. You must select at least one item. |
This page has had
hits since 24 August 1998.
Last updated on 24/08/98