JGAP 3.3.3 released (Java Genetic Algorithms Package)


News: JGAP 3.3.3 released (Java Genetic Algorithms Package)

  1. JGAP is a Genetic Algorithms and Genetic Programming package written in Java. From the JGAP home page:
    WHAT ARE GENETIC ALGORITHMS AND GENETIC PROGRAMS? Genetic algorithms (GAs) are evolutionary algorithms that use the principle of natural selection to evolve a set of solutions toward an optimum solution. GAs are not only quite powerful, but are also very easy to use as most of the work can be encapsulated into a single component, requiring users only to define a fitness function that is used to determine how "good" a particular solution is relative to other solutions. Genetic Programs (GP) enhance GAs. They allow to breed dynamic programs instead of static chromosomes.
    JGAP version 3.3.3 is an extensive release, featuring basically the following:
    • The evolution cycle has been revamped and simplified for a better understanding
    • A lot of smaller enhancements and architectural improvements, see the change log
    • Some bugs have been fixed
    • The basic MinimizingMakeChange example was simplified
    • Improved grid computing by adding new classes and features
    • Javadoc-enhancements
    • New JUnit tests
    This release can be downloaded here: http://sourceforge.net/project/showfiles.php?group_id=11618&package_id=48940 For more information visit the JGAP homepage at http://jgap.sf.net Klaus Meffert for the JGAP team
  2. This sounds interesting, the sort of thing I'd download and play with if I had the time, can anyone give me/us a quick overview of possible applications, ideas etc. to kick start the thought processes please. -John-
  3. Possible use cases[ Go to top ]

    John, Evolutionary Algorithms, which Genetic Algorithms (GA) and Genetic Programming (GP) are part of, are strong in finding near-optimal solutions for problems with a huge solution space. E.g., this includes optimizations or designing circuits. The NASA has evolved an antenna for space that has a superb effectivity. You could uses GAs and GPs to find formulas for a value table. With Genetic Programming in special you could try finding programs that solve (or better: "fit") a given problem. JGAP is used to evolve Java programs that act as robots on the Robocode platform (see http://jgap.sourceforge.net/doc/robocode/robocode.html). Best Klaus http://www.klaus-meffert.com
  4. This algorithm, together with Neural Networks, Baysian Networks, Linear programing are fine solutions I'm always tempted to want to apply in enterprise applications but unfortunately I haven't come across any problem in this space where it could be nicely applied without overkilling the problem. Maybe it's not me but examples that often come with these algorithms. They are so theory-oriented it makes it hard to find a real practical case. That's a pity. Jan
  5. Hi Klaus, There is an area, as you are aware, named Business Intelligence, right? FOr example, we have an application which calculates the incensives a salesmanager gets depending on some rules and some statistics about the sales he had done. We just use a rule engine to get this done. And we have some feature called "what if" analysis, which allows the user to get the details of what he "should" do in order to reach a target. These all are accomplished by some standard rule engines. We also have a reporting platform which generates reports based on Business Objects. What I am not getting or confusing is how a gentic algorthims decisiong is changed from this ? Or in what scenarios I should go for a rule engine like the one above, and where should I be using JGAP? Can you kindly list some scenarios please, thanks Best Ravion