Chess

IdleRich

IdleRich
"In Go it's so much more unclear, making the computer's ability to methodically do hundreds of claculations a second, way more valuable."
But the whole point is that that's obviously not the case cos no computer can get anywhere near a good human player in Go. Number crunching isn't the solution, the computer apparently needs to "think" in a different way and no-one's figured out how to make it do it yet

From wikipedia:

Go poses a daunting challenge to computer programmers. While the strongest computer chess programs can defeat the best human players (for example, the Deep Fritz program, running on a laptop, beat reigning world champion Vladimir Kramnik without losing a single game in 2006), the best Go programs only manage to reach an intermediate amateur level. On the small 9×9 board, the computer fares better, and some programs now win a fraction of their 9x9 games against professional players . Human players generally achieve an intermediate amateur level by studying and playing regularly for a few years. Many in the field of artificial intelligence consider Go to require more elements that mimic human thought than chess.
 

Mr. Tea

Let's Talk About Ceps
If computers do better against good human players on a much smaller board, that would seem to suggest combinatorics is at least part of the problem for AI.
 

IdleRich

IdleRich
From reading further down the page, it is combinatorics that is the problem - for computers at least. Basically in Chess and Go humans are better at dismissing the vast variety of moves and concentrating on the few worthwhile ones. In both games the number of possible moves is hugely greater than the worthwhile ones and as you project several moves forward this discrepancy grows exponentially. However it grows more quickly in Go and although advances in computing have allowed computers to win at chess by sheer number crunching they are still a long way from that in Go. In fact the size of the numbers seems to imply that it would be better to work on building a computer that could figure out how humans did the first bit than to hold out for something that will make Deep Thought look like a toy calculator. It would also be more interesting cos it's always a bit boring to solve something by just going through all the possibilities - I think that's how they proved the 4 Colour Theorem much to many people's disappointment.
 

Slothrop

Tight but Polite
Presumably (from my limited experience) in Go, not only is the branching factor (ie the number of possible moves at any given time) greater, but the number of moves required for the advantage of a particular move to become apparent is greater. Which works out even worse for the computers...
 

4linehaiku

Repetitive
We did some stuff on the differences between Go and Chess in my degree (AI and Computer Science). It's pretty much the number of possible moves leading to huge probability trees as other people have mentioned. Probably as a result of this, the way humans deal with the two problems varies quite a lot. A grand master is supposedly thinking ~12 moves ahead or something, I forget the exact figures (which were obviously fairly vague estimates any way), but the point is they are actually considering all the available moves and plotting the optimal one, though obviously with lots of clever heuristic pruning. So the computer can emulate that style of reasoning pretty well. With Go though that sort of thing hasn't got a chance, and human players rely heavily on pattern matching. Recognising situations that resemble ones in previous games and generalising them mainly. When it comes to that sort of thing I'm afraid the computers are just rubbish.

To continue this branching factor idea, I read a good little anecdote about Marion Tinsley (Best checkers player ever it seems. No, I'd never heard of him either) the other day. He was playing against some computer program and announced "ha you're going to regret that!" after it played a move. His move which capitalised on this mistake was 26 turns later! So I guess that supports the idea that reduced branching allows much deeper search for humans as well.

Oh and I'm shit at chess btw.
 

4linehaiku

Repetitive
I've played that Mancala game as well, though the one my mate had used glass beads not seeds. I wasn't great at that either. Completely trivial for a computer though, there's only n possible moves per go, where n is the number of pits on each side, so 6 - 10.
 

4linehaiku

Repetitive
Finally, the reason computers don't win constantly at bridge / poker etc is because of the hidden information. Can't brute force optimal plays nearly as well when you don't know what your opponent has.
 

baboon2004

Darned cockwombles.
A grand master is supposedly thinking ~12 moves ahead or something, I forget the exact figures (which were obviously fairly vague estimates any way), but the point is they are actually considering all the available moves and plotting the optimal one, though obviously with lots of clever heuristic pruning. .

Apparently, accordsign tot he book I mentioned above, it's much less than people assume (for most grandmasters, sure there're some freakish exceptions), and very few lines are worked out that far in advance unless acutely tactical. general strategic principles take over a lot of the time, which explains why grandmasters still fall for shrouded but not super-complicated tactical ruses sometimes, cos they simply didnt' see any danger and so didn't work out the exact moves.
 

Mr. Tea

Let's Talk About Ceps
Yeah, I also read somewhere that most top players don't actually plan as far ahead as you might assume they do.

I had a conversation with my chess buddy a while back that wandered into general game theory (not a subject I know the first thing about, but anyway...) and for some reason I started wondering if it were theoretically possible to devise a game with rules that are so counterintuitive and seemingly paradoxical - but nonetheless ultimately self-consistent - that someone playing it for the first time would be relatively good but would get progressively worse the more they played it, so that a novice would in general beat a someone who'd played before.

Um, admittedly this conversation happened in a nightclub at around 3am... :slanted:
 

slowtrain

Well-known member
Yeah, I also read somewhere that most top players don't actually plan as far ahead as you might assume they do.

I had a conversation with my chess buddy a while back that wandered into general game theory (not a subject I know the first thing about, but anyway...) and for some reason I started wondering if it were theoretically possible to devise a game with rules that are so counterintuitive and seemingly paradoxical - but nonetheless ultimately self-consistent - that someone playing it for the first time would be relatively good but would get progressively worse the more they played it, so that a novice would in general beat a someone who'd played before.

Um, admittedly this conversation happened in a nightclub at around 3am... :slanted:

I've often wondered (and admittedly, mosttly around sports, which is a bit different) what would happen if you pitted world champions against the worlds WORST players. Would they get completely thrown by the out-of-nowhere, wtf-are-you-doing moves?
 

Slothrop

Tight but Polite
I've often wondered (and admittedly, mosttly around sports, which is a bit different) what would happen if you pitted world champions against the worlds WORST players. Would they get completely thrown by the out-of-nowhere, wtf-are-you-doing moves?

It's not easy to think of sports or games where that would happen.

It's reasonably easy to beat a really bad player at chess, you just follow basic positional principles, be patient and wait for them to balls up.

Total novices can be pretty dangerous at poker, though...
 

Mr. Tea

Let's Talk About Ceps
I've often wondered (and admittedly, mosttly around sports, which is a bit different) what would happen if you pitted world champions against the worlds WORST players. Would they get completely thrown by the out-of-nowhere, wtf-are-you-doing moves?

That's an appealing thought, isn't it? I think it's far more likely the computer would just crank through a very obvious checkmate in four or five moves, however.

You could test it against another AI (or "AS", I suppose!) making moves literally at random, I suppose.

I expect chess ability probably exhibits 'strange loops' on occasion, whereby Player A can usually beat Player B, B can usually beat C but C can usually beat A, because a player's style may have particular strengths in certain areas but idiosyncratic weaknesses in others. Whether the same is true of chess programs, I don't know.

However I'd have thought that most of the time the chess ability hierarchy will be simple and linear, i.e. a player generally regarded as terrible will lose to a merely bad player, the bad player will be beaten by a guy who's kind of OK, and so on up to grandmaster level.
 
Last edited:

Slothrop

Tight but Polite
Apparently, accordsign tot he book I mentioned above, it's much less than people assume (for most grandmasters, sure there're some freakish exceptions), and very few lines are worked out that far in advance unless acutely tactical. general strategic principles take over a lot of the time, which explains why grandmasters still fall for shrouded but not super-complicated tactical ruses sometimes, cos they simply didnt' see any danger and so didn't work out the exact moves.
That makes sense.

This is why I'm always a bit careful talking about this because I don't want to end up with the classic lazy generalisation that chess is simplistic and western and calculating and rational while Go is subtle and eastern and instinctive and, like, totally zen, maaan.

It'd be quite interesting to do brain scans of expert chess and Go players to see whether or not they're using the same parts of the brain when they plan their moves, actually.

But perhaps not interesting enough to justify the effort...
 

baboon2004

Darned cockwombles.
I've often wondered (and admittedly, mosttly around sports, which is a bit different) what would happen if you pitted world champions against the worlds WORST players. Would they get completely thrown by the out-of-nowhere, wtf-are-you-doing moves?

"WHY IS HE HOLDING THE RACKET WITH HIS FOOT??!"

In chess, the simultaneous exhibitions are interesting. Excellent players can crush you (and 40 others at the same time) almost without trying, by never making stupid mistakes.
 
Last edited:

IdleRich

IdleRich
"In chess, the simultaneous exhibitions are interesting. Excellent players can crush you (and 40 others at the same time) almost without trying, by never making stupid mistakes."
I was saying when I was playing against beginners last week that there are two elements that you need to be aware of - one is you have to try and think tactically but the other, arguably more important one, is you have to concentrate and not just forget that something can take your queen. When you start playing you make loads and loads of stupid mistakes and it's easy to forget that once you've stopped doing it.
 

Mr. Tea

Let's Talk About Ceps
That makes sense.

This is why I'm always a bit careful talking about this because I don't want to end up with the classic lazy generalisation that chess is simplistic and western and calculating and rational while Go is subtle and eastern and instinctive and, like, totally zen, maaan.

I can't see how anyone could call chess 'simple'...

I'm sure its, ultimately, possible to compute the optimal Go strategy any number of moves ahead. It's just going to involve a far greater amount of branching than the equivalent number of moves in chess. Where intuition seems to come into it is that human playes use intuition to prune away the overwhelming majority of possible branches that are clearly hopeless, whereas a program (at least a naive, brute-force type program) will mechanically crank through all of them. So in Go you're just going to have to do a lot more pruning - easy for a (good) player, using intuition, but extremely laborious for a computer. But remember that it's only 14 years since Deep Blue stunned the chess worl by beating Kasparov, and now a cheap program that runs on a home PC can reliably beat grandmasters on its hardest setting. So it's only a matter of time before some combination of clever heuristics and sheer CPU power allows a Go program to beat top human players, I reckon.

It'd be quite interesting to do brain scans of expert chess and Go players to see whether or not they're using the same parts of the brain when they plan their moves, actually.

Yeah, for sure!
 
Top