Over to You
You have seen alternative ways to implement the crossover in genetic algorithms and used tournament selection. You also learned about the OneMax problem and elementary cellular automata. You even designed your own CA. You managed to find the best starting cells for the simplest problem, and explored the other setups. As an extension, you can try to find good rules for a fixed starting row. For the ECA you can look things up. It will take a while, but it is possible. Rather than the brute-force approach, you can try a genetic algorithm for this problem too.
You have also seen how machine learning algorithms sometimes fail to solve problems. Trying to model or learn how to improve when the problem is randomly changing is foolish. ...
Get Genetic Algorithms and Machine Learning for Programmers now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
