I have already gone in deep with the GP initialization method and I found that there are some traditional methods that ensure the diversity in the population on the initialization phase of the GP process like RHH, Grow, Full ..etc. my question is if there some other method that ensures the same purpose with those ones or if there some hybridization with other heuristics?
Please read section 3.3.2 of thesis: https://thesis.cust.edu.pk/UploadedFiles/Rohail%20Gulbaz%20(MCS181027).pdf
Check this article as well Article Balancer Genetic Algorithm-A Novel Task Scheduling Optimizat...
Thank you Mohamed Amine Chikh Touami
Hello, I took a look at the article and on the PhD thesis but I guess it talks about Genetic algorithms using a numerical representation of the population.
I'm interested in Genetic programming Tree Representation.
Thank you Rohail Gulbaz
I invite you to read our last publication about improved cuckoo search. The proposed initialization strategy is very efficient to start your search with a good population.
For me, and after the research I did, I guess the best way is to use and go deeper on the Semantic GP, where there are some proposed approaches for maintaining the diversity in a smart way on the initialization step.
- Badges
- Science topic
- Similar topics
- Mathematics
- Algebra
- Matrix