All Classes and Interfaces
Class
Description
Abstract class for population evaluators.
Abstract class for population predictors.
Abstract class for optimization algorithms.
Ant Colony Optimization algorithm.
Compares two arrays of doubles.
BinaryTournament is a selection operator that selects the best individual
from a pair of individuals
BinaryTournamentNSGAII is a selection operator that selects the best individual
from a pair of individuals.
Class to read a BNF grammar from a file.
Boolean mutation operator.
Comparator for NSGA-II algorithm.
This class represents a complex number.
Abstract class for crossover operators.
Assigns the crowding distance to the solutions in the population.
Class for cycle crossover.
This class represents a data table.
Default neighbor generator.
Class implementing the differential evolution technique for problem solving.
Class implementing the differential evolution technique for problem solving.
Random diversification operator.
Abstract class for DTLZ problems
DTLZ1 problem
This class represents the DTLZ1 problem.
DTLZ2 problem
This class represents the DTLZ2 problem.
DTLZ3 problem
This class represents the DTLZ3 problem.
DTLZ4 problem
This class represents the DTLZ4 problem.
DTLZ5 problem
This class represents the DTLZ5 problem.
DTLZ6 problem
This class represents the DTLZ6 problem.
DTLZ7 problem
This class represents the DTLZ7 problem.
EliteSelectorOperator selects the best individuals from the population.
Compares two solutions according to the epsilon-dominance relation.
Evolutionary Strategy algorithm.
This class is used to perform Fast Fourier Transforms.
Extracts the fronts of non-dominated solutions from a population.
Multi-objective Grammatical Evolution Algorithm.
Abstract class for grammatical evolution problems.
Class to represent the phenotype of a GE individual
This class represents a simple example for the Grammatical Evolution algorithm.
Example Please note that using the Script Engine is too slow.
Grammatical evolution using just one objective.
Class to develop "static" (non-temporal) models
This class must be carefully revised and tested.
Class to develop "temporal" models
IntegerFlipMutation mutation operator.
Logger class to setup the logger.
Formatter for the logger.
Master/worker pattern for parallel evaluation of solutions.
This class contains a set of mathematical functions that are used in the
calculation of the statistics of the data.
Mutation operator.
Compiles source and also makes sure that reloading a compiled class
does not "caches" the first compiled class.
Compiles source and also makes sure that reloading a compiled class does not
"caches" the first compiled class.
Class designed to generate neighbors, initially for Tabu Search,
but it can be used for other algorithms.
Assigns the niche count to the solutions in the population.
This class implements a non-uniform mutation operator.
NSGA-II algorithm
Xiaodong Li A Non-dominated Sorting Particle Swarm Optimizer for
Multiobjective Optimization
GECCO-2003, Springer-Verlag, 2003, 2723, 37-48
Input parameters: - C1: PSO c1 factor - C2: PSO c2 factor - CHI: PSO chi
factor - MAX_ITERATIONS - NUM_PARTICLES - SORTING_METHOD: CROWDING_DISTANCE,
CROWDING_DISTANCE_REPLACE, NICHE_COUNT, NICHE_COUNT_REPLACE -
TOP_PART_PERCENTAGE: 0.05 usually, but I am using 0.25 - W: PSO w factor
Compares two solutions according to the value of one of their objectives.
Input parameters: - MAX_GENERATIONS - MAX_POPULATION_SIZE
This class implements a polynomial mutation operator.
Class representing a problem.
Represents a production in a BNF grammar.
Compares two solutions according to the value of one of their properties.
This class provides a set of methods to generate random numbers.
Rastringin benchmark function.
ReductionOperator removes replacementSize Individuals from the population
Enumerates the different types of replacement
Represents a rule in a BNF grammar.
Class for simulated binary crossover.
Scatter Search algorithm
Abstract class for selection operators.
Compares two solutions according to the value of their first objective.
This class implements a simple genetic algorithm.
Class implementing the simulated annealing technique for problem solving.
Class for single point crossover.
Class representing a solution in a problem.
Compares two solutions according to the dominance relation.
Class representing a set of solutions in a problem.
SPEA2 algorithm
This class contains methods to manage strings.
SwapMutation mutation operator.
Class designed to generate neighbors.
Class to represent a symbol in a grammar
Enum to represent the type of the symbol
Tabu search algorithm.
TournamentSelect is a selection operator that selects the best individual
from a set of individuals.
Traveling Salesman Problem.
This class implements the unary hypervolume indicator as proposed in Zitzler,
E., and Thiele, L. (1998): Multiobjective Optimization Using Evolutionary
Algorithms - A Comparative Case Study.
UniformMutation mutation operator.
Class representing a variable in a problem.
Worker thread for parallel evaluation of solutions.
Abstract class for ZDT problems
ZDT problems are a family of multi-objective optimization problems.
ZDT1 benchmark problem
ZDT1 is a multi-objective optimization problem
f1(x) = x1
f2(x) = g(x) * h(f1(x), g(x))
g(x) = 1 + 9 * (sum(xj) / (n-1))
h(f1, g) = 1 - sqrt(f1 / g)
xj in [0, 1]
Pareto optimal front is a convex curve
ZDT2 benchmark problem
ZDT2 is a multi-objective optimization problem
f1(x) = x1
f2(x) = g(x) * h(f1(x), g(x))
g(x) = 1 + 9 * (sum(xj) / (n-1))
h(f1, g) = 1 - (f1 / g)^2
xj in [0, 1]
Pareto optimal front: convex
ZDT3 benchmark problem
ZDT3 is a multi-objective optimization problem
f1(x) = x1
f2(x) = g(x) * h(f1(x), g(x))
g(x) = 1 + 9 * (sum(xj) / (n-1))
h(f1, g) = 1 - sqrt(f1 / g) - (f1 / g) * sin(10 * pi * f1)
xj in [0, 1]
Pareto optimal front: non-convex
ZDT4 benchmark problem
ZDT4 is a multi-objective optimization problem
f1(x) = x1
f2(x) = g(x) * h(f1(x), g(x))
g(x) = 1 + 10 * (n-1) + sum(xj^2 - 10 * cos(4 * pi * xj))
h(f1, g) = 1 - sqrt(f1 / g)
x1 in [0, 1]
xj in [-5, 5]
Pareto optimal front: non-convex
ZDT6 benchmark problem
ZDT6 is a multi-objective optimization problem
f1(x) = 1 - exp(-4 * x1) * sin(6 * pi * x1)^6
f2(x) = g(x) * h(f1(x), g(x))
g(x) = sum(xj) / (n-1)
h(f1, g) = 1 - (f1 / g)^2
xj in [0, 1]
Pareto optimal front: non-convex