/Simulation_experiments_for_optimizing_objective_function

Simulation experiments for optimizing objective function with Differential Evolution, Evolution Strategies and Cross Entropy Method (2 versions)

Primary LanguageJupyter NotebookMIT LicenseMIT

Simulation experiments for optimizing objective function

Table of contents

  1. Prerequisites
  2. Introduction
  3. Differential Evolution
  4. Evolution Strategies
  5. Cross Entropy Method
  6. Improved Cross Entropy Method

1. Prerequisites

pip install -r requirements.txt

2. Introduction to objective function for optimization

a. Sphere function:

Screen Shot 2021-10-26 at 19 59 54
Screen Shot 2021-10-26 at 19 59 06

b. Rosenbrock function

Screen Shot 2021-10-26 at 20 01 39
Screen Shot 2021-10-26 at 20 03 03

c. Rastrigin function

Screen Shot 2021-10-26 at 20 05 34
Screen Shot 2021-10-26 at 20 05 57

d. Griewank function

Screen Shot 2021-10-26 at 20 08 12
Screen Shot 2021-10-26 at 20 09 18

e. Ackley function

Screen Shot 2021-10-26 at 20 12 50
Screen Shot 2021-10-26 at 20 13 46

3. Differential Evolution

a. Sphere

Sphere-DE-1024

b. Rosenbrock

Rosenbrock-DE-512

c. Rastrigin

Rastrigin-DE-512

d. Griewank

Griewank-DE-32

e. Ackley

Ackley-DE-1024

4. Evolution Strategies

a. Sphere

Sphere-ES-1024

b. Rosenbrock

Rosenbrock-ES-1024

c. Rastrigin

Rastrigin-ES-1024

d. Griewank

Griewank-ES-1024

e. Ackley

Ackley-ES-1024

5. Cross Entropy Method

a. Sphere

Sphere-CEM-1024

b. Rosenbrock

Rosenbrock-CEM-1024

c. Rastrigin

Rastrigin-CEM-1024

d. Griewank

Griewank-CEM-1024

e. Ackley

Ackley-CEM-1024

6. Improved Cross Entropy Method

a. Sphere

Sphere-CEMv2-1024

b. Rosenbrock

Rosenbrock-CEMv2-1024

c. Rastrigin

Rastrigin-CEMv2-1024

d. Griewank

Griewank-CEMv2-1024

e. Ackley

Ackley-CEMv2-1024