Using c#,c/c++ or java to improve BBN with GA

Question

I have a little problem in my little project , I wish that someone here could help me!

I am planning to use a bayesian network as a decision factor in my game AI and I want to improve the decision making every step of the way , anyone knows how to do that ? Any tutorials / existing implementations will be very good,I hope some of you could help me.

I heard that a programmer in this community did a good implementation of this put together for poker game AI.I am planning to use it like him ,but in another poker(Texas) or maybe Rentz.

Looking for C/c++ or c# or java code.

Thanks , Mike

Answer 1

You can try Charniak's Bayesian Networks Without Tears. For Bayesian Network implementations, look at BUGS and LibB.

Answer 2

There are two steps:

1. You need to know how to apply a Bayesian Network and how to define the nodes and the belief propagation for your game. To do this, you need to read tutorials.

2. Apply software. This is quite easy, there are plenty of free/open source implementations. At the end of the wiki page http://en.wikipedia.org/wiki/Bayesian_network, there are a list of software. You may want to check Infer.Net, which is a mature library developed in MSR Cambridge.