GA is a way to learn a population by generations, better understand a problem and working towards a goal.
This by giving a small fragments of the both parents dna (information/mutation) into the childs “dna”.
For each new generations, they will given more understanding of the problem and thereby make better actions towards the problem.
Only the best children are allowed into the next generation with some miner mutation, to remove the section where they could be quite monotonous otherwise.
Here are some links for more information:
Wiki about Genetic Algorithm
AI-Depot (a good AI page with many other solutions of AI stategy)
I have added a small Tetris application where GA is used to learn the generation to play better. A small data-file is added, so they will not have to start from scratch.
Quite interesting how the generation works 🙂
Limitations: No limitations