RE: [Bug-gnubg] Neurogammon by Tesauro

[Alain Redlinger]

Hi Ian,
I do not have the papers you mention, but you might have a look at these:

http://www.research.ibm.com/massive/tdl.html#h1:temporal_difference_learning

http://satirist.org/learn-game/systems/gammon/

best regards,
Alain 

-----Message d'origine-----
De : address@hidden
[mailto:address@hidden De la part de Ian Shaw
Envoyé : vendredi 19 juin 2009 16:47
À : bug-gnubg
Objet : [Bug-gnubg] Neurogammon by Tesauro

I'm trying to read up on the creation of the hand-crafted Neural Net
Inputs.

Does anyone have a copy of any of the following papers, which they'd be
willing to share?

Neurogammon: A Neural-Network Backgammon Program
By: G. Tesauro
Published in: IJCNN 90 International Joint Conference on Neural
Networks, vols.1-3. , New York, IEEE, p.C33-40 in 1990


Neurogammon Wins Computer Olympiad
By: Gerald Tesauro
Neural Computation Fall 1989, Vol. 1, No. 3, Pages 321-323

I already have the following, which gives some good explanation, but I'd
like to see what Tesauro has to say on the matter.

BKG A Program That Plays Backgammon
Berliner
1977

Regards,
Ian Shaw

 


_______________________________________________
Bug-gnubg mailing list
address@hidden
http://lists.gnu.org/mailman/listinfo/bug-gnubg