chess - exact science?
Tim923
inspiration, and creativity, or can someone with enough intelligence,
such as a math PhD wizard, learn it as a science and be a grandmaster?
Are the grandmasters outstanding mathematicians?
Sterten
>inspiration, and creativity, or can someone
probably some amount of all
with enough intelligence,
>such as a math PhD wizard, learn it as a science and be a grandmaster?
yes, but the question is how much time and
energy will it take ?
In theory you can learn it as a science, but the rules for good play or
positional understanding are not clearly defined.
There are just too many possible situations
to be subsumed into position-classes.
One difference is, that at chess you have to be quick and concentrated.
In math you can make pauses,or do something irrelevant, even make mistakes-
there is a chance to correct them later.
So maybe math is a bit more like correspondence chess ?!?
>
>Are the grandmasters outstanding mathematicians?
I think, there is a greater ratio of mathematicians among grandmasters than
among normal people.
See e.g.:
http://cadigweb.ew.usna.edu/~wdj/math_chess.htm
Keith A. Lewis
>Does being super great at chess require unteachable talent,
>inspiration, and creativity, or can someone with enough intelligence,
>such as a math PhD wizard, learn it as a science and be a grandmaster?
Are you saying that one can become a "math PhD wizard" without natural
talent, inspiration, or creativity?
I think the aptitude for abstract reasoning is a requirement to excel in
either area. In my high school the chess club and math team shared a lot of
members.
>Are the grandmasters outstanding mathematicians?
The renaissance people among them, maybe. Most probably specialize in one
or the other.
--Keith Lewis klewis {at} mitre.org
The above may not (yet) represent the opinions of my employer.
Augustus S.F.X Van Dusen
I actually hold top chess players in rather low regard, intellectually
speaking. My belief is that they are little more than idiot-savants. A
brute force approach to chess can already beat them anyway - but no
computer can do mathematics the way event the humblest of mathematicians
can.
Dave Seaman
Emanuel Lasker, a former world chess champion, was a math professor. In
fact, he introduced the notion of a primary ideal and proved the primary
decomposition theorem for an ideal of a polynomial ring in terms of
primary ideals.
--
Dave Seaman
Judge Yohn's mistakes revealed in Mumia Abu-Jamal ruling.
<http://www.commoncouragepress.com/index.cfm?action=book&bookid=228>
Randy Poe
Datapoint of one: I've always been decent in math and science,
got A's in undergrad and grad courses, work in the sciences
professionally.
Terrible at chess. Terrible at card games. Terrible at
war games. Terrible at any game of strategy. I just simply seem
to have no mind for analyzing positions or thinking about
strategies. I can do OK in backgammon, but that's about it.
Even checkers seems to defeat me.
One of my brothers is an MD. Another is a writer, mostly
in business and politics. Neither one considers math to be a
forte. Both were always excellent at all of the above.
- Randy
Leonard Blackburn
> On Wed, 04 Aug 2004 00:48:19 +0000, Tim923 wrote:
>
> > Does being super great at chess require unteachable talent,
> > inspiration, and creativity, or can someone with enough intelligence,
> > such as a math PhD wizard, learn it as a science and be a grandmaster?
> >
> > Are the grandmasters outstanding mathematicians?
>
> I actually hold top chess players in rather low regard, intellectually
> speaking. My belief is that they are little more than idiot-savants.
I don't think your belief is correct. I have followed the careers of
many top chess players, having read many interviews and articles, and
I believe most of them are quite well-rounded and very intelligent
people. Also, I believe that making it to the top in chess requires
intelligence, discipline, creativity, hard-work, good health, etc. I
don't think it requires some freakish natural ability to calculate
deeply and quickly, for example. One great chess champion was once
asked how many moves does he think ahead in a typical position. His
response (without too much exaggeration) was "one." Although most top
players are amazing calculators and can even play blindfolded (but
these are, I believe, learned skills).
> A
> brute force approach to chess can already beat them anyway
the fact that human chess players don't play by brute force weakens
your stance that the best human chess players are like idiot savants.
> - but no
> computer can do mathematics the way event the humblest of mathematicians
> can.
Humans can also beat the best computers (just not every time). In
fact, when
the best humans don't make tactical (calculational) errors and are
playing their best they usually make the best computers look like
idiots. Computers are very poor at strategy and in some positions
they can't see as far ahead as humans can. Computers lack the
intuition and creativity of the top human chess players.
Now the fact that the gap between human mathematicians and computer
mathematicians is much greater than the gap between human chess
players and computer chess players does not say anything about human
chess players versus human mathematicians. Rather, it says something
about chess versus mathematics.
-Leonard
Augustus S.F.X Van Dusen
> Also, I believe that making it to the top in chess requires
> intelligence, discipline, creativity, hard-work, good health, etc. I
> don't think it requires some freakish natural ability to calculate
> deeply and quickly, for example.
I agree about the hard work and discipline sides. My understanding is
that top players have an obsessive dedication to the game - similar to
what idiot-savants do in their own domains.
> Humans can also beat the best computers (just not every time).
At this stage everything seems to be in place to build machines that
would beat any human, all the time, consistently. It's little more than a
matter a brute force. Anyway, I grant that this is beside the point, for
it is evident that humans play chess in a totally different way. What is
not beside the point is that they are better at it - just like cars are
better at 100 meters sprints than people.
> Computers lack the intuition and creativity of the top human chess
> players.
I wonder about that? The fact is, we don't know how chess players do
their thing. For all we know, computations similar to those carried out in
a computer might be taking place in their brains at a subliminal level -
which might be described as "intuition" and "creativity" at the
conscious level.
> Now the fact that the gap between human mathematicians and computer
> mathematicians is much greater than the gap between human chess players
> and computer chess players does not say anything about human chess
> players versus human mathematicians. Rather, it says something about
> chess versus mathematics.
I would have thought that, in part, what happens is that chess is a
finite, albeit large, game, with a fixed, unambiguous set of rules -
whereas mathematics is an infinite discipline, where new rules are
invented all the time. This is why I don't think much of chess and chess
players, intellectually spaeking, and why I believe that chess playing
machines will eventually become unbeatable by any human.
Let me put it in a different way: to me, top chess players are a bit like
Johann Zacharias Dase. Interesting, even intriguing, attractive for show
biz purposes, and little more. Actually, Dase was more useful in his time.
David Bandel
> > mathematicians is much greater than the gap between human chess players
> > and computer chess players does not say anything about human chess
> > players versus human mathematicians. Rather, it says something about
> > chess versus mathematics.
>
> I would have thought that, in part, what happens is that chess is a
> finite, albeit large, game, with a fixed, unambiguous set of rules -
> whereas mathematics is an infinite discipline, where new rules are
> invented all the time. This is why I don't think much of chess and chess
> players, intellectually spaeking, and why I believe that chess playing
> machines will eventually become unbeatable by any human.
you talked yourself into the same mistake you made in your first
point.
it says something about chess vs. mathematics which means you're
opinion of chess players, intellectually speaking is poorly founded.
and for your extremely, poor information, most top grandmasters have
genius IQ's and perfect memories. most world champions were prolific
mathematicians.. those that weren't were brilliant musicians which we
all know go hand in hand with mathematicians. just because a genius
takes up chess instead of mathematics does not take away from the fact
that he's a genius.
in conclusoin, you're dumb and everyone thinks so and i hope you burn
in a hell inviting buffoons like you sit at the right hand of the
devil, ocasionaly leaning over to suck his burning cock.
Michael Jørgensen
Well, the Norwegian Grandmaster Simen Agdestein is also a professional
soccer player. Not sure if that is relevant to the discussion, though :-)
-Michael.
Michael Jørgensen
> inspiration, and creativity, or can someone with enough intelligence,
> such as a math PhD wizard, learn it as a science and be a grandmaster?
Well, like most other aspects of life, you need both talent and hard work.
Hard work alone will only get you half the way. In my experience, anyone
with enough determination could just about manage to hit the bottom of the
world ranking list (ELO rating of 2000). Anything beyond that will require
some measure of talent.
The same applies to mathematics. You might earn good grades in graduate
school from sheer hard work, but anything close to a professional occupation
requires some skill/flair/talent.
-Michael.
Augustus S.F.X Van Dusen
> and for your extremely, poor information, most top grandmasters have
> genius IQ's and perfect memories.
Assuming that to be true, that doesn't necessarily imply intellectual
prowess. There is more to that than just perfect memory and high IQ - as
Marilyn Vos Savant so perfectly illustrates, who with a purported IQ of
200, intellectually speaking seems to have been able to do little more
than writing an amusing column in an insignificant weekend magazine, and
one or two misguided books on mathematics.
It was Enrico Fermi (an intellectual giant) who said that Nobel prize
laureates (many of them intellectual giants) had very little in common,
even in intelligence.
> most world champions were prolific mathematicians..
Maybe. However, thinking about a few outstanding, recent ones (Kasparov,
Fisher, Spassky, Petrosian, Karpov, Korchnoi) I am not aware that they
have contributed anything to mathematics. I think that Karpov used to be
an economist, but I am also not aware that he had any influence in that
discipline.
> those that weren't were brilliant musicians which we
> all know go hand in hand with mathematicians. just because a genius
> takes up chess instead of mathematics does not take away from the fact
> that he's a genius.
Of course. You can also be a genius at juggling oranges, driving cars,
golfing, playing poker, etc., which doesn't automatically turn you into an
intellectual overachiever.
> in conclusoin, you're dumb and everyone thinks so and i hope you burn
> in a hell inviting buffoons like you sit at the right hand of the
> devil, ocasionaly leaning over to suck his burning cock.
:-) :-) :-)
I'll leave the entelechy of hell for you alone. Oops! I am sorry! You
probably don't know what an entelechy is anyway.
Leonard Blackburn
>
> > Also, I believe that making it to the top in chess requires
> > intelligence, discipline, creativity, hard-work, good health, etc. I
> > don't think it requires some freakish natural ability to calculate
> > deeply and quickly, for example.
>
> I agree about the hard work and discipline sides. My understanding is
> that top players have an obsessive dedication to the game - similar to
> what idiot-savants do in their own domains.
Some, but not all are obsessive. Kasparov is very politically active.
Morozevich gave an interview last year expressing that chess is just
not a feasible career for him anymore. I've heard lots of
grandmasters say that
they don't put in more than a few hours a week studying chess.
Usually, the chess greats start when they are young children. They
make a choice of a career very early. I don't blame them. They find
out that they are extremely good at something, that they can win
trophies, money, praise, and fame. Why not give it a shot as a
career? It's like a sport. I love mathematics, but if I were as good
at golf as Phil Michelson, I'd be right out there on that pga tour.
I'd turn math into a serious hobby instead of a career. I wouldn't do
the same with chess, though, since there's only one chess player in
the world (Kasparov) guaranteed of being rich, and you need to be a
top twenty player to even make a living playing chess.
The bottom line is, a choice of a career doesn't automatically make
one any more or any less intelligent.
>
> > Humans can also beat the best computers (just not every time).
>
> At this stage everything seems to be in place to build machines that
> would beat any human, all the time, consistently. It's little more than a
> matter a brute force. Anyway, I grant that this is beside the point, for
> it is evident that humans play chess in a totally different way. What is
> not beside the point is that they are better at it - just like cars are
> better at 100 meters sprints than people.
>
> > Computers lack the intuition and creativity of the top human chess
> > players.
>
> I wonder about that? The fact is, we don't know how chess players do
> their thing.
Actually, we do know a lot about it. The top chess players publish
detailed analysis of their own games, often giving deep explanation of
how they came to certain moves. There are also many books written by
grandmasters that explain their approach to playing chess.
But my statement about computers lacking intuition and creativity is
vague.
There are actual concrete things we can point to that computers are
lacking
in chess playing ability. For example, I remember a recent match
between
Kasparov and Deep Junior (maybe Deep Fritz) in which Kasparov won a
game that made the computer look foolish because a certain key
strategic plan was beyond its brute force search horizon. Basically,
in a quite closed position, Kasparov had complete control of the queen
side and had forseen that he should open up the queen's rook file and
break through. But it took him about twenty moves to implement this
strategy. Meanwhile, what the computer should have been doing was
pushing its king-side pawns for counterplay (instead of making useless
waiting moves). But the computer would not do so because it gives
poor marks to exposing his own king, and the plusses that would have
compensated for these poor marks were too many moves ahead for the
computer to search.
Basically, computers cannot see general characteristics of a chess
position and plan strategies. They really can't foresee positions
unless they have calculated a path to them. But their search
algorithms are very sophisticated and based on actual positional chess
principles. Humans can visualize where they would like their pieces
to be without actually calculating how they would get there. Humans
can also plan very long-range strategic goals, something current
computers cannot do. That is why the top humans can still make
computer programs look foolish from time to time. When computers beat
the
top ten players in the world it is usually because they out-calculate
them.
All that said, the calculational power of chess engines is becoming so
great and sophisticated that occasionally they surprise humans by
making very human-like positional moves.
> For all we know, computations similar to those carried out in
> a computer might be taking place in their brains at a subliminal level -
> which might be described as "intuition" and "creativity" at the
> conscious level.
I think if one learns a lot about chess and chess players, then one
realizes that this is certainly not what's going on. Besides it just
seems like a very far-fetched guess.
>
> > Now the fact that the gap between human mathematicians and computer
> > mathematicians is much greater than the gap between human chess players
> > and computer chess players does not say anything about human chess
> > players versus human mathematicians. Rather, it says something about
> > chess versus mathematics.
>
> I would have thought that, in part, what happens is that chess is a
> finite, albeit large, game, with a fixed, unambiguous set of rules -
> whereas mathematics is an infinite discipline, where new rules are
> invented all the time. This is why I don't think much of chess
That, I understand.
> and chess
> players,
That, I don't.
> intellectually spaeking, and why I believe that chess playing
> machines will eventually become unbeatable by any human.
>
I think that even the top ten chess players in the world agree with
you on that last point.
> Let me put it in a different way: to me, top chess players are a bit like
> Johann Zacharias Dase. Interesting, even intriguing, attractive for show
> biz purposes, and little more. Actually, Dase was more useful in his time.
-Leonard
Sterten
I think, this is all due to the annoying and time-comsuming programming
problems
and that in theory _any_ skills and
superiorities of humans in chess vs. computers
can be also programmed for computers
and eventually will be done one day
and by then you will no longer be able
to detect any situation where humans
will be better than computers.
It's just all so difficult to implement and why
waste so much time in these details when
computer speed is still advancing with
Moore's law and just waiting makes
them stronger far more than including
one more of these human-made details ?
I think,it is possible to write a chessprogram
of ELO3000 on a 1MHz machine with 1MB
of memory in theory.
--Guenter.
Stephen
"Keith A. Lewis" <le...@PROBE.mitre.org> a écrit dans le message de
news:ceqp2u$bfi$1...@newslocal.mitre.org...
> I think the aptitude for abstract reasoning is a requirement to excel in
> either area.
Sure, but getting it is only a matter of experience...
Michael Jørgensen
"Sterten" <ste...@aol.com> wrote in message
news:20040805125046...@mb-m02.aol.com...
> reading here a lot about computer weaknesses at chess ..
>
> I think, this is all due to the annoying and time-comsuming programming
> problems
> and that in theory _any_ skills and
> superiorities of humans in chess vs. computers
> can be also programmed for computers
> and eventually will be done one day
> and by then you will no longer be able
> to detect any situation where humans
> will be better than computers.
>
> It's just all so difficult to implement and why
> waste so much time in these details when
> computer speed is still advancing with
> Moore's law and just waiting makes
> them stronger far more than including
> one more of these human-made details ?
I agree with you.
> I think,it is possible to write a chessprogram
> of ELO3000 on a 1MHz machine with 1MB
> of memory in theory.
Well, here I simply disagree. Much of the human strength comes from the
ability to do *parallel* processing. This is also called pattern
recognition, and results is an statement like "My opponents kingside attack
is gaining too much momentum. I need to create counterplay now". Yes, this
could be programmed into a computer, but it is just too complex to describe
in a *sequential* programming language.
Humans are also able to adapt their play according to their opponent. If I
play against a very week player, I usually just shuffle my pieces
back-and-forth, waiting to him to make a serious mistake. This is far
*safer* for me, than to take any unnecessary risks by plunging forward with
a desperate attack. However, if my opponent is equal or better than me, then
I'm forced to take more risks. This really boils down to a boolean decision:
Should I defend or should I attack? Well, the answer depends on my knowledge
of my opponent.
However, this is all moot. My guess is that the increased computing power
alone will eventually wipe out all human competition, given another 20 years
or so.
-Michael.
David Bandel
well what's your point? are you saying you don't respect all highly
intelligent people solely on the basis of how they choose to use it?
if one person does applied mathematics and another does pure
mathematics.. u'd consider the pure mathematician more "intelligent"
than the other? your argument isn't making any sense.. not to mention
using words i don't understand (like entelechy)
Joseph Hertzlinger
<blac...@math.umn.edu> wrote:
> But my statement about computers lacking intuition and creativity is
> vague. There are actual concrete things we can point to that
> computers are lacking in chess playing ability. For example, I
> remember a recent match between Kasparov and Deep Junior (maybe Deep
> Fritz) in which Kasparov won a game that made the computer look
> foolish because a certain key strategic plan was beyond its brute
> force search horizon. Basically, in a quite closed position,
> Kasparov had complete control of the queen side and had forseen that
> he should open up the queen's rook file and break through. But it
> took him about twenty moves to implement this strategy. Meanwhile,
> what the computer should have been doing was pushing its king-side
> pawns for counterplay (instead of making useless waiting moves).
> But the computer would not do so because it gives poor marks to
> exposing his own king, and the plusses that would have compensated
> for these poor marks were too many moves ahead for the computer to
> search.
I'm reminded of "64-Square Madhouse" by Fritz Leiber
eagleso...@yahoo.com
The number of possible games is to large to speculate. So
each move is either aan attack a defense or both.
And the move is based on a special speculative games held
secret. One for the defender and one for the player.
And here is the question in another form. Should the
contest be allowed passive play until a mistake is made?
A standoff is easy to maintain. And recognizing the mistake
is the easy way to win.
So the judge does not exist like in boxing, where boring
contests are sometimes disallowed. The referee says,
"pick it up. please" or somebody losses a point.
A grandmaster has a personality of the chess master
at an early age. To learn to compete with one is not
very smart, but doable.
Start to play to gain experience is the required
method of learning the art of chess.
A science approach to winning is really cheating, because
of the parity playing required.
Except I never play chess, because of this reason.