[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [gnugo-devel] what method is used in AI for playing go

From: Shreya Inamdar
Subject: Re: [gnugo-devel] what method is used in AI for playing go
Date: Thu, 9 Aug 2012 19:13:37 +0530

GNUGo has Monte Carlo method already implemented (Along with being an expert system).If you want to base your work on GNUGo, the to do list is a great place to start with.

If you are planning to do AI around go, I suggest you get a very raw implementation of GO bot. There's the wikibook on it.Then you can add your own modules.

Proving results is a bit tricky in GO implementation.True, there are servers for competing, but it takes a lot of time.So, first be clear on how the results will be shown.

On Thu, Aug 9, 2012 at 6:49 PM, Petri Pitkanen <address@hidden> wrote:
reading the documentation would have answered your question. But Gnu is largely an expert system. Obviously go needs tree search for local skirmishes as well


2012/8/8 Erick Morales <address@hidden>
Good day,

I'm right now working in my thesis work to obtain the title of "system engineer", so I'm have interest on doing some AI around go.

I am thinking in using swarm intelligence.

So, I will appreciate any help about how GNUGo works, for example, ¿Do you use Monte Carlo Method? if not, Artificial Neural Networks?


gnugo-devel mailing list

gnugo-devel mailing list


reply via email to

[Prev in Thread] Current Thread [Next in Thread]