On Fri, 11 Jul 2003, Neil Fernandez wrote:

How would people rate the relative strengths of the best programs at
chess, Chinese chess, and go?

I realise that it's not a straightforward task to find a basis for
comparison (including because the processing power made available to
chess programs when playing Kramnik and Kasparov has not [?] been made
available to Chinese chess and go programs), but I'd be interested in
what people think :-)

Hi Neil,

The algorithms for playing chess are more straightforward than any
developed so far for Go. Go is also much bigger in terms of possible
moves, significance of positions, etc. Chess programs have evolved to
the point where the possibilities can be looked up and number crunched
to the level of a grandmaster.

Go isn't anywhere near that stage and all programs developed so far play
at a relatively 'experienced beginner' level. It's not for lack of
processing grunt, it's down to Go needing human intelligence and
intuition to play well.

You might find this site interesting

http://www.intelligentgo.org/en/comp.../overview.html

--
Chris

For computer,

Grand master (chess), street-corner player (chinese chess, Shogi), beginner
(Go).

Chinese Chess is on 9 by 10 board and Shogi on 9 by 9 board. Both can be
solved the chess way - (I think)

Go is so complicated that computer power is not a major factor. The need
for power grow exponentially for linear increase in strength.

james

"Neil Fernandez" wrote in message
...
How would people rate the relative strengths of the best programs at
chess, Chinese chess, and go?

I realise that it's not a straightforward task to find a basis for
comparison (including because the processing power made available to
chess programs when playing Kramnik and Kasparov has not [?] been made
available to Chinese chess and go programs), but I'd be interested in
what people think :-)

Another version of the question would be: how do the three games compare
with each other in terms of how difficult it is to tell a computer how
to play strongly.

Neil

--
Neil Fernandez

In rec.games.go Chris Lawrence wrote:

: Go isn't anywhere near that stage and all programs developed so far play
: at a relatively 'experienced beginner' level. It's not for lack of
: processing grunt, it's down to Go needing human intelligence and
: intuition to play well.

I believe this is a bit premature of a statement. Simply because
processing power currently is not good enough to play a good game of go
does not mean that at some point in the future it won't be. We (humans)
play to a large degree by "human intelligence" and "intuition" because we
also lack that processing power. There is no reason to assume that once
the processing power exists, a computer will not emerge to challenge the
best.

I do realize the massive amounts of positions, et cetera, that need to be
considered, but it seems entirely possible that algorithms might emerge in
the future which can do large overviews of the board and return ideas
which to a human might pass for intuition and might avoid the raw
computation of the possibility-trees altogether.

Best regards,
Justin

"james" wrote:

Go is so complicated that computer power is not a major factor.

Currently available hardware/software does not scale well. There has been
hardly any improvement in playing strength since HandTalk 1996.
The board size is not a major factor either. The best available software do
not perform any better (and often worse) on 9x9 than on 19x19 boards.

--
JVT

Dear Justin,

: Go isn't anywhere near that stage and all programs developed so far play
: at a relatively 'experienced beginner' level. It's not for lack of
: processing grunt, it's down to Go needing human intelligence and
: intuition to play well.

I believe this is a bit premature of a statement. Simply because
processing power currently is not good enough to play a good game of go
does not mean that at some point in the future it won't be. We (humans)
play to a large degree by "human intelligence" and "intuition" because we
also lack that processing power. There is no reason to assume that once
the processing power exists, a computer will not emerge to challenge the
best.

As Chris suggested, processing power isn't the only question. How we
program the computer also matters. And, perhaps, computer architecture.

The brain is massively parallel. It can do serial processing, but
computer programs beat it all hollow in that regard. It may be that
advances in programming for parallel processing will be more important
than increases in processing power.

Best,

Bill

In article gers.com,
james writes

For computer,

Grand master (chess), street-corner player (chinese chess, Shogi),

I'd be surprised (and very interested!) if that were the case for
Chinese chess, at least when the best programs are run with the same
processing power at which chess programs play at GM level.

I'm a lot weaker at Chinese chess than I am at chess (I've spent much
less time at it), but it seems to me to be a less complicated game than
chess. It's also more 'tactical' - pawns capture as they move and
therefore can't block each other, and the game is much more open.

beginner (Go).

Chinese Chess is on 9 by 10 board

True but apart from pawns, only 4 out of each player's 9 pieces can
cross the river. Of the other 5, 2 can reach only 7 squares, and 3 are
restricted to a 3x3 area, in which 2 of them can reach only 5 squares.

Neil

--
Neil Fernandez

In article , Bill Spight
writes

Dear Justin,

: Go isn't anywhere near that stage and all programs developed so far play
: at a relatively 'experienced beginner' level. It's not for lack of
: processing grunt, it's down to Go needing human intelligence and
: intuition to play well.

I believe this is a bit premature of a statement. Simply because
processing power currently is not good enough to play a good game of go
does not mean that at some point in the future it won't be. We (humans)
play to a large degree by "human intelligence" and "intuition" because we
also lack that processing power. There is no reason to assume that once
the processing power exists, a computer will not emerge to challenge the
best.

As Chris suggested, processing power isn't the only question. How we
program the computer also matters. And, perhaps, computer architecture.

The brain is massively parallel. It can do serial processing, but
computer programs beat it all hollow in that regard. It may be that
advances in programming for parallel processing will be more important
than increases in processing power.

In addition it would be wrong to assume that there is no game with a
clearly defined and finite state space at which no computer programs
will be _able_ to play as strongly as the strongest humans can, even
with lots of processing power and parallel processing.

Neil

--
Neil Fernandez

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Justin O. Wyss-Gallifent wrote:
In rec.games.go Chris Lawrence wrote:

: Go isn't anywhere near that stage and all programs developed so far play
: at a relatively 'experienced beginner' level. It's not for lack of
: processing grunt, it's down to Go needing human intelligence and
: intuition to play well.

I believe this is a bit premature of a statement.

Oh I just classified it in with the statement by Pachman about computers
never reaching Grandmaster level in chess because ...

Whilst humans can do it and computers can't it is called intelligence
and intuition**, and when the computers overtake us it is called brute
force, number crunching, rule following etc.

Humans don't like the idea of computers being "better" than they are, so
we tend to judge them by the things we are good at, playing Go,
translating poetry, and talking. Rather than the things they are good
at, like chess, maths, logic, data handling.

It is of course an irrational response, as we are nothing like
computers, so trying to draw a comparison is pointless. A program either
is or isn't better at a specific task than a specific human, assuming
that that task requires special human like powers, unless you have
evidence to that effect, other than no one has figured out how best to
program it, is also irrational.

I do realize the massive amounts of positions, et cetera, that need to be
considered, but it seems entirely possible that algorithms might emerge in
the future which can do large overviews of the board and return ideas
which to a human might pass for intuition and might avoid the raw
computation of the possibility-trees altogether.

I'd say probable, as far as we can ascertain massively parallel
computers have exceeded the processing and storage capacities of the
human brain, so all we need is the software for whatever task is at
hand, be it Go, or consciousness (whatever that may be).

I see no evidence from biology, chemistry or physics, that human brains
do anything "weird", "mysterious" or "quantum" (as certain maths
Professors might have us believe).

I suspect the primary reason that computers beat the world at Chess and
not Go, are cultural, the west followed this route in AI and used chess
as its guinea pig, the east seemed more interested in robots and ended
up with large amounts of the worlds car manufacturing industry. If
Shannon and Turing had been more into Go than Chess, I wonder what
modern computers would be like?

If I had to stick my neck out I'd say, Go may be the first game where
the heuristics used to defeat the best humans are computer generated.
Automated weighting adjustments were needed for both Checkers and Chess,
but I suspect Go may be mastered by a program where that "tuning" or
"evolution" is taken to the point where the "authors" don't understand
how the program is choosing it's moves.

Simon

** I just finished ploughing through the analysis of a load of chess
games, where again and again human "intuition" produced better moves
than the computer analysing the games, yet the program is demonstrably
much stronger than any of the human players involved.
-----BEGIN PGP SIGNATURE-----
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQE/DuotGFXfHI9FVgYRAskgAKC9lgzFd/oCWyZtYwfSHSG6YBYyUQCgjDEp
bPDe6EBhgKc8Kfp5WzATv50=
=M7O7
-----END PGP SIGNATURE-----

In article , JVT
writes

The best available software do
not perform any better (and often worse) on 9x9 than on 19x19 boards.

So is it basically the case that the best available go program, given
the most processing power it has ever run on, would lose every single
game against a weak club-level human opponent on a 9x9 board?

Neil

--
Neil Fernandez

In article , Neil Fernandez wrote:
In article , JVT
writes

The best available software do
not perform any better (and often worse) on 9x9 than on 19x19 boards.

So is it basically the case that the best available go program, given
the most processing power it has ever run on, would lose every single
game against a weak club-level human opponent on a 9x9 board?

followups set to rgg

I'd expect to win in excess of 80% on 9x9 against any of the top PC
programs on 9x9, assuming reasonable komi[1]. I might do better than
that.

For what it's worth, I'm a British 3 kyu, which isn't *weak* - in chess
terms, somewhere between 1500 and 1700 ELO, I'd speculate (judging by
how difficult it's felt to get to this strength in both games; I was a
little under 2000ELO at chess when I was playing regularly.)

I'm certainly not strong at Go, though. [2]

- Andrew

[1] compensatory points for going second
[2] Yet... (I hope! )

--
Andrew Walkingshaw |