Assignment 1

group 59: Adrian, Caterina, Kamiel, Katja

useful links:

[https://github.com/DEAP/deap]

Next meetings:

Friday, 18.09.2020 at 14:00 - TSH

Cate's algorithm:

Features:

• random uniform weights (not changed)
• discrete recombination (page 65)
• random resetting mutation (page 56)
• steady-state model (page 80): the best half of the population mate and the next generation will be half made by the parents, half made by the offspring (mutated) --> maybe it's a slow algorithm?

Possible changes or improvements:

• different distribution of the weights
• blend or arithmetic recombination
• nonuniform mutation
• different mating pool: generational model
• allowing duplicates between the parents
• deciding the dimension of lambda (offspring) and mu (parents)