Every story and comment from the Hacker News API, in a layout borrowed from 2001.
A comment on Introduction to Writing Genetic Algorithms.
Population size 1024 took only 52 generations to get from random gibberish to "Helln, woqld!"
Then over 6000 generations to arrive to "Hello, world!", which seems not very efficient.
As every member of population is evaluated for fitness every generation, we performed 6M evaluations.
If same task was bruteforced (all combinations of 13 length from alphabet of 89 letters 89^13 variants) it would take much more than 6M.
There are no comments on this comment yet.
Population size 1024 took only 52 generations to get from random gibberish to "Helln, woqld!"
Then over 6000 generations to arrive to "Hello, world!", which seems not very efficient.
As every member of population is evaluated for fitness every generation, we performed 6M evaluations.
If same task was bruteforced (all combinations of 13 length from alphabet of 89 letters 89^13 variants) it would take much more than 6M.