[cs250] theory2

[Sean McCoy]

Hi Jay,

Here's theory2.

Thanks,

Sean

[jame...@vassar.edu]

This is really interesting. I knew that a lot of research went into AI for chess, but I had no idea that the same went for checkers and connect four. It makes you wonder if there is a solution for everything in life given a strong enough computer (unless life is NP complete). I found the idea of endtables interesting because when I wrote my Monte Carlo Search Tree for Prog2, I figured that there was probably a lot of wasted computation because there are a bunch of ways to get to the same move. It's a little sad though. Kind of takes the fun out of games. Having a computer beat a person in a game doesn't seem as impressive if the computer basically has a solution to the game. I guess that's why we need more complex games like Go, where computers can only think so far in advance.