• A new LGBTQ+ forum is now being trialed and there have been changes made to the Support and Advice forum. To read more about these updates, click here.
  • Hey Trainers! Be sure to check out Corsola Beach, our newest section on the forums, in partnership with our friends at Corsola Cove! At the Beach, you can discuss the competitive side of the games, post your favorite Pokemon memes, and connect with other Pokemon creators!
  • Due to the recent changes with Twitter's API, it is no longer possible for Bulbagarden forum users to login via their Twitter account. If you signed up to Bulbagarden via Twitter and do not have another way to login, please contact us here with your Twitter username so that we can get you sorted.

Crunching the numbers: Game trees: Enhance your game with the power of math

Status
Not open for further replies.

BulbaBot

Dreams of electric Bulbasaur
Joined
Nov 13, 2005
Messages
8,546
Reaction score
178
Crunching the numbers: Game trees: Enhance your game with the power of math

Back in 1996, a computer defeated a world champion chess player at a single round of chess for the first time. How did it do this? By computing every possible move and working out which one was best. This may same incredibly inefficient, but it payed off. How does this apply to Pokémon? And how can humans apply this method? Danielle Detering investigates.

Read more on Bulbanews
 
Neat read. Being interested in both AI and Pokemon, I find the game theory behind the games fascinating. Quite complex for a "children's game".

Anyway, at the end the author asks what allows people to remain competitive against computers in chess. The answer is that people are much better at pattern recognition than computers. Chess masters are familiar with hundreds of different board patterns and sub-patterns and they know good moves to make in these situations. Even though every game of chess is unique, the same general strategies show up repeatedly. Because of this, people don't have to analyze millions of moves. By recognizing these patterns, they can greatly prune the move tree and only have to analyze a small subset of all the possible moves.
 
I was worried the crunching the numbers series was dead. I'm glad it isn't, the articles are really interesting.
 
As my programming mentor says, computers are quite stupid, in fact. They are just efficient at crunching numbers. This is what binary code and other programming is. It's basically a language of number crunching to the computer, and then it reads the code. All computers can do is take orders and crunch numbers.
 
Great article! It caught my eye when you talked about chess, and then you went on to the actual subject, which was just as interesting.
 
This was a really awesome Article
Pokemon, at its roots feels very simple
However, with 6 different choices and 4 different moves for each choice and on top of that weather and abilities

These charts become all the more complicated, though built up from very simple things

I really appreciated your article and I hope to read more from you in the future :)
 
Aww! Thanks for all the positive comments guys. It's the little things in life that keep me going.

What about paralysis? :p

I'm glad someone noticed that the model isn't perfect yet. I plan to explain chance events in detail eventually, but if you want to get a head start on the subject, you can look up "Extensive-form games".
 
Status
Not open for further replies.
Back
Top Bottom