Hello everyone. Here is Mingxiao Lyu. Simply call me Ming.
It seems this title is a little big. But it doesn’t matter, just a few personal thoughts. It starts at yesterday when I watched a video on some Chinese video site like YouTube. It was a lecture about game balance. You know I like playing video games and that’s why I watched that video. But the thing is that in that video the talker talked about Zermelo’s theorem in game theory, says that in any finite two-person game of perfect information in which the players move alternatingly and in which chance does not affect the decision making process, if the game cannot end in a draw, then one of the two players must have a winning strategy.
That sounds crazy because according to this theorem, players in the Chess, or in the Go must have a winning strategy. And if so why people still play these games and hold the international matches for them? Although it seems unbelievable, it is the truth. Because Zermelo’s theorem has been proved by Mathematical induction. The games like Chess or Go do have a winning strategy, but people can’t find it. The reason is that there are too many situations in these games. For example, there are about 2.08 x 10^170 situations in the Go (from the research by Princeton University). That number is bigger than the number of atoms in our known universe. Since we can’t figure out all of these situations we can’t find that strategy to make sure our winning.
Then a thought came out that if people can’t find that strategy, how about computers? In recent years people gradually find out that computers can do so many works for us. And the answer to that question, which people thought is “no” before, may change to “yes” in the future not so far. People ever thought that computers must be limited by it’s speed of calculation and can’t figure out games like Go forever. As I said before that there are about 2.08 x 10^170 situations in the Go since the fastest computer in the world now can only reach about 10^20 times calculation per second, which is not enough at all. So people believed that computer can’t defeat us in the Go in 10 years. However, actually they did.
Lee Sedol played an historic five game match against Google DeepMind’s AlphaGo computer program in March 2016. AlphaGo won the match, making it the first time a computer Go program had defeated a world class human player on even terms: AlphaGo 4 — Lee Sedol 1