# Algorithms

## Classes

class  AlgAEliGpea
Asynchronous elimination global parallel algorithm. More...
class  ArtificialBeeColony
Artificial Bee Colony algorithm (see e.g. http://www.scholarpedia.org/article/Artificial_bee_colony_algorithm)ABC algorithm accepts only a single FloatingPoint genotype (vector of real values). Additionally, it adds the following genotype for algorithm implementation:

• FloatingPoint genotype as trial (a cycle counter for each individual).
More...
class  Clonalg
Clonal Selection Algorithm (see e.g. http://en.wikipedia.org/wiki/Clonal_Selection_Algorithm).This CLONALG implements:

• cloning Versions:
• static cloning : n of the best antibodies are cloned beta*populationSize times
• proportional cloning: number of clones per antibody is proportional to that ab's fitness
• inversely proportional hypermutation : better antibodies are mutated less
• selectionSchemes:
• CLONALG1 - at new generation each antibody will be substituded by the best individual of its set of beta*population clones
• CLONALG2 - new generation will be formed by the best (1-d)*populationSize clones ( or all if the number of clones is less than that )
• birthPhase: where d * populationSize of new antibodies are randomly created and added to the population for diversification.
More...
class  CuckooSearch
Cuckoo search (CS) optimization algorithm (see http://en.wikipedia.org/wiki/Cuckoo_search)CS algorithm accepts only FloatingPoint genotype (vector of real values). This implementation is based on: http://www.mathworks.com/matlabcentral/fileexchange/29809-cuckoo-search--cs--algorithm. More...
class  DifferentialEvolution
Differential evolution (DE) optimization algorithm (see e.g. http://en.wikipedia.org/wiki/Differential_evolution)DE algorithm accepts only a single FloatingPoint genotype (vector of real values). More...
class  Elimination
Elimination (generation gap) algorithm with roulette wheel elimination selection operatorThis algorithm is genotype independent (it can be used with any Genotype). More...
class  EvolutionStrategy
(mu/rho +/, lambda) - Evolution Strategy (ES) algorithm.This algorithm is genotype independent (it can be used with any Genotype). More...
class  GeneticAnnealing
Genetic annealing algorithm (see e.g. http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.56.7606, http://drdobbs.com/architecture-and-design/184409333?pgno=10)Currently implemented only for minimization problems! More...
class  OptIA
Optimization Immune Algorithm (opt-IA, see e.g. http://www.artificial-immune-systems.org/algorithms.shtml).This opt-IA implements:

• static cloning: all antibodies are cloned dup times, making the size of the clone population equal dup * poplationSize
• inversely proportional hypermutation: better antibodies are mutated less
• static pure aging - if an antibody exceeds tauB number of trials, it is replaced with a new randomly created antibody
• birthPhase: if the number of antibodies that survive the aging Phase is less than populationSize, new randomly created abs are added to the population
• optional elitism.
More...
class  Algorithm
Algorithm base class. More...
class  ParticleSwarmOptimization
Particle swarm optimization algorithm (see e.g. http://en.wikipedia.org/wiki/Particle_swarm_optimization)PSO algorithm accepts only a single FloatingPoint genotype (vector of real values). Additionally, it adds the following genotypes for algorithm implementation:

• FloatingPoint genotype as particle velocity
• FloatingPoint genotype as best-so-far position
• FloatingPoint genotype as best-so-far fitness value.
More...
class  RandomSearch
Random search algorithmThe algorithm flow: More...
class  RouletteWheel
Generational algorithm with roulette wheel selection operatorThis algorithm is genotype independent (it can be used with any Genotype). More...
class  AlgSGenGpea
Synchronous generational global parallel algorithm. More...