ec.multiobjective.README Maven / Gradle / Ivy

Go to download

Show more of this group Show more artifacts with this name
Show all versions of ecj Show documentation

ECJ, A Java-based Evolutionary Computation Research System. ECJ is a research EC system written in Java. It was designed to be highly flexible, with nearly all classes (and all of their settings) dynamically determined at runtime by a user-provided parameter file. All structures in the system are arranged to be easily modifiable. Even so, the system was designed with an eye toward efficiency. ECJ is developed at George Mason University's ECLab Evolutionary Computation Laboratory. The software has nothing to do with its initials' namesake, Evolutionary Computation Journal. ECJ's sister project is MASON, a multi-agent simulation system which dovetails with ECJ nicely.

The newest version!

This package, and various subpackages, contain tools for doing multiobjective
optimization.  In the primary package (ec.multiobjective) are just useful
top-level classes.  In the subsidiary packages (ec.multiobjective.nsga2
and ec.multiobjective.spea2) are implementations of two popular multiobjective
optimization algorithms, NSGA-II and SPEA2, which use these top-level classes.




The top-level classes:


ec.multiobjective.MultiObjectiveFitness

This is the superclass of fitness classes involving multiple objectives.  This
class holds not a single fitness value but some N objective values.  These 
objective values can be either maximized or minimized depending on the setting
of the fitness class.  The class can perform Pareto Dominance computation between
two MultiObjectiveFitness classes, and in fact uses this as the basis of its
default fitness comparison.  Additionally the class contains utility methods for
computing Pareto Fronts.


ec.multiobjective.MultiObjectiveStatistics

This simple subclass of ec.simple.SimpleStatistics aso outputs the Pareto Front
in various ways at the end of the run:

	- A summary printed on-screen in a helpful columnar format
	- The entire front written to the end of the statistics file
	- The multiple objective results of the front written to a special
	  auxillary file (nominally called 'front.stat').  This front can
	  be easily displayed by gnuplot, Excel, or other plotting packages.  
	  If you have a single subpopulation, for example, you can write:

		echo plot \"front.stat\" | gnuplot -persist

	  and the front will be displayed in gnuplot for you.