FW: [Bug-gnubg] Training gnu backgammon

[Joachim Matussek]

Hello,

i have read several times that there are new efforts to train new neural nets 
for GNUBG. I believe there is a high risk of wasting gazillions of CPU cycles 
if you donŽt have a very good concept.

In my opinion you need at least one person who is a strong BG player and one 
person who is expert in training neural nets (especially backgammon neural 
nets). They have to go through several steps if they want to succeed in 
improving GNUBG. I will write down some of the most important items.

0) Read all of Hans Berliner on BKG.
1) Analyze the strengths and weaknesses of the existing GNUBG neural nets and 
the partitioning of the backgammon position space within GNUBG (e.g. 
weaknesses: containment play/ almost all of crashed net/ odd-even-ply bias...)
2) Think about an improved partitioning of the backgammon position space. Think 
of a quick algorithm how to decide which neural net is suited for a particular 
position.
3) Think about an improved coding of the backgammon board (raw and additional 
inputs) depending on the position type.
4) Analyze the former training process of GNUBG (detailed documentation 
required). Decide what the best training process will be (TD training doesnŽt 
give very accurate neural net but is able to learn from scratch/ supervised 
training gives accurate nets only if the training data are very good)
5) Do some experiments on the size of the neural nets (accuracy vs. speed). 
DonŽt forget problems like overfitting and generalisation.
6) Start TD training for parts of the game where GNUBG is too weak to use the 
existing rollout or ply data.
7) Acquire training data for supervised training by rollouts and ply 
evaluations.
8) Start supervised training -> 9.
9) Test the resulting neural nets -> 8.

Have fun,

Joachim Matussek


> -----Ursprüngliche Nachricht-----
> Von: "Christian Anthon" <address@hidden>
> Gesendet: 10.03.07 17:50:52
> An: "Joseph Heled" <address@hidden>, bug-gnubg <address@hidden>
> Betreff: [Bug-gnubg] Training gnu backgammon
Hi Joseph,
> 
> could you tell(remind) us how to get started using gnubg-nn for training of 
> the neural nets. What I would like to know is
> 
> A) how is it decided which net (normal/race/crashed) that is used.
> B) are the crashed and normal nets different in their design or are they just 
> trained against different sets of positions
> 
> C) how to train a net by self-play (TD training?)
> D) how to train a net against a database of positions
> 
> Christian.
> 
> -----------------------------------------------------------------
> _______________________________________________
> Bug-gnubg mailing list
> address@hidden
> http://lists.gnu.org/mailman/listinfo/bug-gnubg
> 


_____________________________________________________________________
Der WEB.DE SmartSurfer hilft bis zu 70% Ihrer Onlinekosten zu sparen!
http://smartsurfer.web.de/?mc=100071&distributionid=000000000066