org.moeaframework.algorithm
Class VEGA

java.lang.Object
  org.moeaframework.algorithm.AbstractAlgorithm
      org.moeaframework.algorithm.AbstractEvolutionaryAlgorithm
          org.moeaframework.algorithm.VEGA

All Implemented Interfaces:

Algorithm, EvolutionaryAlgorithm

public class VEGA
extends AbstractEvolutionaryAlgorithm

Implementation of the Vector Evaluated Genetic Algorithm (VEGA). VEGA should be avoided in practice, since many modern algorithms outperform it and exhibit better convergence properties, but is included due to its historical significance. VEGA is considered the earliest MOEA. It supports M objectives during the selection phase by selecting M different subgroups, each selected based on the i-th objective value, for i=1,...,M.

There is one small algorithmic difference between this implementation and [1]. In [1], applying the genetic operators fills the entire population. However, since custom variation operators can be specified, it is possible that the population will not be filled completely. As a result, this implementation will continue selecting parents until the population is full.

References:

1. Schaffer, D. (1985). Multiple Objective Optimization with Vector Evaluated Genetic Algorithms. Proceedings of the 1st International Conference on Genetic Algorithms, pp. 93-100.

Field Summary

Fields inherited from class org.moeaframework.algorithm.AbstractEvolutionaryAlgorithm

archive, initialization, population

Fields inherited from class org.moeaframework.algorithm.AbstractAlgorithm

initialized, numberOfEvaluations, problem, terminated

Constructor Summary

VEGA(Problem problem, Population population, NondominatedPopulation archive, Initialization initialization, Variation variation)
Constructs a new VEGA instance.

Method Summary

protected void

iterate() Performs one iteration of the algorithm.

Methods inherited from class org.moeaframework.algorithm.AbstractEvolutionaryAlgorithm

getArchive, getPopulation, getResult, getState, initialize, setState

Methods inherited from class org.moeaframework.algorithm.AbstractAlgorithm

evaluate, evaluateAll, evaluateAll, finalize, getNumberOfEvaluations, getProblem, isInitialized, isTerminated, step, terminate

Methods inherited from class java.lang.Object

clone, equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Methods inherited from interface org.moeaframework.core.Algorithm

evaluate, getNumberOfEvaluations, getProblem, isTerminated, step, terminate

Constructor Detail

VEGA

public VEGA(Problem problem,
            Population population,
            NondominatedPopulation archive,
            Initialization initialization,
            Variation variation)

Constructs a new VEGA instance.

Parameters:

problem - the problem

population - the population

archive - the external archive; or null if no external archive is used

initialization - the initialization operator

variation - the variation operator

Method Detail

iterate

protected void iterate()

Description copied from class: AbstractAlgorithm

Performs one iteration of the algorithm. This method should be overridden by implementations to perform each logical iteration of the algorithm.

Specified by:

iterate in class AbstractAlgorithm