[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
brute force and knowledge
Dan Liu wrote :
>I agree with some of the posts that search (brute or not) is very important in
>computer Go. Actually it's the only hope that a Go program can win over a high
>ranking professional player. The reasons are the following. ..
John Clarke wrote :
>But what is the human brain if not the archetypical "brute force" machine?
>Neurons aren't very smart - there are just lots of them.
>
>Almost certainly humans use a massively parallel search to choose their moves.
>
>Of course we both agree that humans only evaluate high depth trees to *check*
>possible plays, and even those trees are severely pruned.
>
>But the basic algorithm is surely just a very big, very parallel search. And
>the more search the better.
Darren Cook wrote :
>If brute force means doing lots and lots of simple tasks - tedious and
>stupid by human standards - then I think when very strong go programs
>appear they will be doing brute force. Of course it is likely to be a
>different kind of brute force from that used in chess (exhaustive search to
>10-20 ply).
OK, seems we're always back to the same old questions and topics.
Seems that these times we have a lot of brute force defenders ! Strange for the
computer game that is supposed to be the most untractable by brute force !
I'd like to add some personal thoughts :
1) I said that go will never be solved by brute force : of course this is
speculative, and I was referring to the high branching factor of go, nothing
more. But I think erveryone would agree that if brute force is to be used in go,
then the evaluation fonction has to be *very* smart (may be using "internal
regularities")
2) how could you say that, in the end, human brain uses brute force ? Yes, it is
highly parallel. But that's necessary, not enough, for the brain to be like a
massive parallel computer. It might be so, but it might not : maybe there is
some "global" complex structures and behaviour that arise when you put together
an enough number of basic devices (such as neurons), but maybe also there is a
"soul" that is NOT localized in the brain and that need only a brain to express
itself ?
3) This topic is also higly philosophical. I have one opinion, you may have an
other one, but for me the essence of go is in the question : why are human go
experts so unable to express their knowledge ?
just (for me) because go is more artistic than mathematic : go is a writing that
need to people to be put on the (paper) board. Go is a painting. Go is a
building of a new world at each new game, and maybe even our universe is a
gigantic go game ! (This is serious, since there is a new phys theory called
ElectroMagnetic Quantum Gravity, or EMQG, compatible with relativity, and based
on the idea that our world is really a cellular automata). Yes, playing go is
like creating a new space-time.
4) thoses questions come "exponentially" when the board size increase. A
computer can play well up to 7x7 by using brute force only. A child of age 8 can
play very well on a 9x9. On a 13x13, you have still to be able to balance local
and global play, and maybe there is already "something" at this size that cannot
be captured by a machine. On a 19x19, the game is so complex that you can work
hard for years and play just 1-dan in the end : the existence of higher rank
players is a complete mystery. And I think that capturing the true "essence" of
a 25x25 game is beyond human capabilities ! Not to speak of an infinite game
(the mathematical study of it certainly would be interresting !)
5) Furthermore, strong human players have all their own philosophy of what go
is, and they know that their opponent has its too, and they play accordingly.
The stones position on the board is not the only "input " for humans !
I do not think that computer go is hopeless, but I think we have a lot of
progress to do in AI, philosophy and even maths before we are able to build a >
5 dan computer go player... Unless we find a breackthrough idea that use somehow
the mathematical properties of go.
My own program "Spirit Of Go" has NO pattern recognition modules, NO expert
system, NO genetic
algorithm, and definitively NO evaluation function nor search function ! How
then could it work ?
Well, well, youl'll see at next Ing Cup, I hope...
--
______________________
/ Let java be with me !\ \\\|///
\______________________/ O \\ ^ ^ //
o o ( @ @ )
+--------------------------------------oOOo-(_)-oOOo----+
| Serge Boisse |
| SERVICE TECHNIQUE DE LA NAVIGATION AERIENNE (STNA) |
| PHIDIAS project, http://www.stna.dgac.fr/phidias |
| tel: (33)562 14 5731 |
| mailto:boisse@xxxxxxxxxxxxxxxxx |
| homepage: http://www.multimania.com/boisse |
+-----------------------------------------------Oooo----+
oooO ( )
( ) ) /
\ ( (_/
\_)