Applied Computational Intelligence and Soft Computing / 2010 / Article / Alg 1

Research Article

Adaptive Representations for Improving Evolvability, Parameter Control, and Parallelization of Gene Expression Programming

Algorithm 1

ACS mutation operator pseudocode
Data: Chromosome
Result: Mutated chromosome
  /*Calculate the decay rate     */
  decayRate = 1 (gen+maxGen*factor)/maxGen
  /*Calculate the mutation rate,
   inverse to the fitness      */
  muP = (1-chr.Ftn/bestFtn)
  /*Adjust the mutation rate if it is
   below the minimum        */
  if     then
   muP = minRate
  /*Apply the decay to the mutation
   rate              */
  muP = muP * decayRate
  /*Determine if mutation will occur
  if     then
   /* Randomly decide to grow or
    shrink             */
   growChromosome = DoCoinToss()
   if     then
     /*Grow the chromosome by
      adding a new gene     */
     insertionPoint = GetRnd(0, chr.NGenes)
     /*Shrink the chromosome by
      deleting a gene, but only
      if we have at least two
      genes            */
     if     then
      deletionGene = GetRnd(1,