PV226 ML: Differential Evolution

Content of this session

Explanation of Differential Evolution

Diffent approches to candidate calculation

And development

Differential Evolution

works best on real numbers

does not require continuous space

on those cases works better than GA

from wiki

Base Vector Options

Random
Best
Current

(drawing placeholder)

Differential Evolution in Sklearn

from scipy.optimize import differential_evolution
import numpy as np
def ackley(x):
    arg1 = -0.2 * np.sqrt(0.5 * (x[0] ** 2 + x[1] ** 2))
    arg2 = 0.5 * (np.cos(2. * np.pi * x[0]) + np.cos(2. * np.pi * x[1]))
    return -20. * np.exp(arg1) - np.exp(arg2) + 20. + np.e
bounds = [(-5, 5), (-5, 5)]
result = differential_evolution(ackley, bounds)
result.x, result.fun

Any questions?

What to add to the mix

Coevolution