Evolutionary

A Julia package for evolutionary & genetic algorithms.

Documentation	Build Status	References

Installation

The package can be installed with the Julia package manager. From the Julia REPL, type ] to enter the Pkg REPL mode and run:

pkg> add Evolutionary

Algorithms

• (μ/ρ(+/,)λ)-SA-ES (ES)
• (μ/μ_I,λ)-CMA-ES (CMAES)
• Genetic Algorithms (GA)
  • Multi-objective: NSGA-II (NSGA2)
• Differential Evolution (DE)
• Genetic Programming (TreeGP)

Operators

• Mutations

  • ES
  • (an)isotropic Gaussian
  • (an)isotropic Cauchy
  • GA
  • binary
    • flip
    • inversion
  • real valued
    • uniform
    • Gaussian
    • BGA (domain range)
    • (MI)PM (power)
    • PLM (polynomial)
  • combinatorial
    • inversion
    • insertion
    • swap2
    • scramble
    • shifting

• Recombinations

  • average
  • marriage

• Crossovers

  • binary
  • SPX (single point)
  • TPX (two point)
  • SHFX (shuffle)
  • UX (uniform)
  • BINX (binary)
  • EXPX (exponential)
  • BSX (binary subset)
  • real valued
  • DC (discrete)
  • AX (average)
  • WAX (weighted average)
  • IC (intermediate)
  • LC (line)
  • HX (heuristic)
  • (MI)LX (Laplace)
  • SBX (simulated binary)
  • combinatorial
  • PMX (partially-mapped)
  • OX1 (order)
  • OX2 (order-based)
  • CX (cycle)
  • POS (position-based)

• Selections

  • rank-based fitness assignment
  • (μ, λ)-uniform ranking
  • roulette (fitness proportionate selection, FSP)
  • stochastic universal sampling (SUS)
  • tournament
  • truncation