problems that solve themselves #shorts

Key Takeaways

have you all ever heard of genetic algorithms they're basically this sort of magical way of solving complex problems where the problem actually just solves itself they're actually modeled after the way all living things evolve through natural selection all little gray dots you see are a hundred mice right now they have zero idea where they're going but after the first generation only the top fifty percent of mice survive based on how much food they got from their environment that means the next generation is a little bit better adapted to know where the food is and as time goes on we eventually arrive at a generation of mice which can always find a sufficient supply of food so why don't we just use this to solve all of our problems well as you might have noticed there's no guarantee of finding the best possible solution in this next simulation we take the environment to be a little bit more harsh so that only the top 25 percent survive they still settle on the plus one at the end of the day but take fewer generations to find it and finally here's a really harsh environment where only the top five percent of mice survive each generation and almost magically they're able to find that plus 10 best solution in no time at all