Re: computer-go: Learning from existing games

[schraudo@xxxxxxxxxxx]

On Tue, 14 Jan 2003, Heikki Levanto wrote:

> On Mon, Jan 13, 2003 at 03:36:55PM -0800, Piotr Kaminski wrote:
> > Nicol Schraudolph et al.:  Learning to Evaluate Go Positions Via Temporal
> > Difference Methods (http://www.inf.ethz.ch/~schraudo/pubs/gochap.pdf)
> 
> Interesting paper. I had not thought that a neural network that just sees
> the board position would ever get that far.

Thanks Heikki!  I should add that this work was 10 years ago and could
certainly be done much better now.  Having said that, TD-learning is
clearly not *the* solution to computer go (as it was for backgammon).
Used correctly, it can be a powerful tool though.

> Most current programs extract a lot of information from the position
> (grouping of stones into strings, strings into groups, counting liberties,
> eyes, and territory, and so on). I feel that this sort of info would be much
> better input for a network that will estimate the winning probability (or
> score).

Agreed.  Markus Enzensberger has been doing very interesting work in that
direction.  Markus, if you're listening - how is NeuroGo doing these days?

> Actually, I believe it would be enough to count a pile of key numbers that
> reflect the position. Number of captured stones, number of strings and
> stones in atari, with 2-5 liberties; number and total size of living groups,
> weak groups, dead groups; number of points under more or less strong
> black/white control; and so on. A small set of (say) 100 numerical inputs
> ought to suffice. I believe TD-learning would work better on such a network.

I suspect that throwing away all spatial information will cost you, at
least in that the program won't learn to play good shape.  Perhaps combine
it with a local shape-based candidate move generator (which could also be
a neural network)?

Best,

- Nici.

-- 
    Dr. Nicol N. Schraudolph             http://www.inf.ethz.ch/~schraudo/
    Inst. of Computational Science               mobile:  +41-76-585-3877
    ETH Zentrum, HRS H30                            tel:      -1-251-3661
    CH-8092 Zuerich, Switzerland                    fax:        -632-1374