Generalized Generation Gap with Parent-Centric Recombination (G3PCX)

class pypop7.optimizers.ga.g3pcx.G3PCX(problem, options)[source]

Generalized Generation Gap with Parent-Centric Recombination (G3PCX).

Note

Originally G3PCX was proposed to scale up the efficiency of GA mainly by Deb, the recipient of IEEE Evolutionary Computation Pioneer Award 2018.

Parameters:

  • problem (dict) –

    problem arguments with the following common settings (keys):

    • ’fitness_function’ - objective function to be minimized (func),

    • ’ndim_problem’ - number of dimensionality (int),

    • ’upper_boundary’ - upper boundary of search range (array_like),

    • ’lower_boundary’ - lower boundary of search range (array_like).

  • options (dict) –

    optimizer options with the following common settings (keys):

    • ’max_function_evaluations’ - maximum of function evaluations (int, default: np.Inf),

    • ’max_runtime’ - maximal runtime to be allowed (float, default: np.Inf),

    • ’seed_rng’ - seed for random number generation needed to be explicitly set (int);

    and with the following particular settings (keys):

    • ’n_individuals’ - population size (int, default: 100),

    • ’n_parents’ - parent size (int, default: 3),

    • ’n_offsprings’ - offspring size (int, default: 2).

Examples

Use the optimizer to minimize the well-known test function Rosenbrock:

 1>>> import numpy
 2>>> from pypop7.benchmarks.base_functions import rosenbrock  # function to be minimized
 3>>> from pypop7.optimizers.ga.g3pcx import G3PCX
 4>>> problem = {'fitness_function': rosenbrock,  # define problem arguments
 5...            'ndim_problem': 2,
 6...            'lower_boundary': -5*numpy.ones((2,)),
 7...            'upper_boundary': 5*numpy.ones((2,))}
 8>>> options = {'max_function_evaluations': 5000,  # set optimizer options
 9...            'seed_rng': 2022}
10>>> g3pcx = G3PCX(problem, options)  # initialize the optimizer class
11>>> results = g3pcx.optimize()  # run the optimization process
12>>> # return the number of function evaluations and best-so-far fitness
13>>> print(f"G3PCX: {results['n_function_evaluations']}, {results['best_so_far_y']}")
14G3PCX: 5000, 0.0

For its correctness checking of coding, refer to this code-based repeatability report for more details.

n_individuals

population size.

Type:

int

n_offsprings

offspring size.

Type:

int

n_parents

parent size.

Type:

int

References

https://www.egr.msu.edu/~kdeb/codes/g3pcx/g3pcx.tar (See the original C source code.)

https://pymoo.org/algorithms/soo/g3pcx.html

Deb, K., Anand, A. and Joshi, D., 2002. A computationally efficient evolutionary algorithm for real-parameter optimization. Evolutionary Computation, 10(4), pp.371-395. https://direct.mit.edu/evco/article-abstract/10/4/371/1136/A-Computationally-Efficient-Evolutionary-Algorithm