On Tue, May 15, 2012 at 7:22 PM, Craig Weinberg <whatsons...@gmail.com>wrote:

>
> > Deep Blue has several possible moves
> > and chooses one of them (just as Kasparov does). It makes a decision each
> > move. And given that it eventually gets to check-mate, Deep Blue wins
> chess.
>
> That's only the view of a human being who is familiar with chess.


I agree that we are not talking about frogs watching chess games. But
a human being watching the match will see that Deep Blue makes decisions
and wins the game.


Deep
> Blue is neither a human or familiar with chess.

When you add 5+6 into
> a calculator, it does not 'decide' that the answer is 11 any more than
> a square peg decides it doesn't fit in a round hole.

If Deep Blue had
> a perspective, which it doesn't, it would have no idea who Kasparov is
> or that he was the opponent. No clue that check-mating Kasparov is
> good or that being check-mated is bad. The game of chess is in the eye
> of the beholder, not in the computation of statistics.
>
>
To me, a good decision in the context of chess is that which allows to win
a chess game. Everything else is pretty irrelevant. What is your definition
for a good chess decision?

>
> > Deep Blue decides what piece to move and where to move it.  That counts
> as
> > a decision to me.
>
> I understand that, I'm just trying to tell you why that doesn't work.
> Deep Blue decides nothing.


We use Deep Blue to inform us what the most
> mathematically efficient chess move is and then we can choose to
> imagine that we are playing a game against an entity that is deciding
> to make those moves. There is no entity there though. The computer is
> a puppet.
>

There are two entities there. One is Kasparov and the other one is Deep
Blue. Both of them decide what pieces to move. In fact, they move them.
Nobody is imagining anything. That is what we see.

You can win chess by looking at every possible outcome of every
> possible move and putting them in order of how few moves will likely
> end the game in your favor. There is no decision at all, you are just
> organizing a stack of finite patterns in order of probable efficiency.
>

That's a decision to me: several alternatives and the ability to rank them.
if you had complete information, that's how you should decide things.
Should a human being do otherwise if he had perfect knowledge?


> There is nothing to decide, you just solve the math problem and report
> the result as your move.


The problem is that in the case of chess, the math problem cannot be solved
exactly, not even close, with the resources available currently (probably
never). Both Kasparov and Deep Blue must resort to heuristics, previous
knowledge, and learning. Deep Blue also loses games, it has not perfect
knowledge.

Ricardo.

-- 
You received this message because you are subscribed to the Google Groups 
"Everything List" group.
To post to this group, send email to everything-list@googlegroups.com.
To unsubscribe from this group, send email to 
everything-list+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/everything-list?hl=en.

Reply via email to