Re: [asa] NASA - Climate Simulation Computer Becomes More Powerful

Iain wrote:
> Although you maintain the answer is "no" that computers would out-do
> humans in mental capacity, if the answer turned out to be "yes" it
> would be an immediate and complete refutation of Dawkins's famous
> "Ultimate Boeing 747 gambit" argument - that the creator must be more
> complex than that which is created - presumably we would have created
> the computer that outdoes us.
>
> In a weaker sense, the existence of computer chess programs like "Deep
> Blue" that beat Kasparov (often thought to be the greatest player
> ever) has already refuted the Dawkins complexity argument. The
> programmers of Deep Blue know a lot less about playing chess than does
> Kasparov, and if they played against their creation it would crush
> them far more easily than it crushed Kasparov.

I wouldn't be so quick to say that Deep Blue refuted Dawkins's complexity
argument, even in a weaker sense. Actually, the subject of Deep Blue brings
back a lot of memories. I became VP of science and technology, which
included the Deep Blue team, the very week that Deep Blue first played
Kasparov and he won. I had the privilege of hosting him at a luncheon
afterwards. He was in a great mood and it was a lot of fun. In a move of
corporate politics, a colleague of mine managed to wrest the Deep Blue group
away from me before the final contest when Deep Blue won. (In hindsight, I
was very glad, seeing the media turmoil) Kasparov was not in such a good
mood after that. But in the short time I had the team, I learned a lot about
how Deep Blue was architected and programmed. Your last sentence is
certainly correct--Deep Blue beat everyone else more easily than Kasparov.
But Deep Blue isn't necessarily more complex than any of its creators. Deep
Blue was designed with a set of accelerator chips built specifically for
chess. The real challenge was in the programming. We hired a grandmaster who
worked with the team incessantly, playing chess over and over. They tuned
the weighting parameters to a great precision. That of course is subjective
and reflects the style of play of that grandmaster.

The complexity of Deep Blue was greater than its creators in the sense of
being able to calculate an enormous number of possible scenarios and
calculate all their weighting values. No human can do that. But humans have
a much better intuitive sense of figuring it out without such brute force
calculations. That's a complexity no computer can touch.

There was a good reason why we decided not to pursue a rematch. There was no
benefit for the company or for learning about computer technology by just
doing more of the same. The point had been made, to great fanfare, and that
was enough.

Randy

To unsubscribe, send a message to majordomo@calvin.edu with
"unsubscribe asa" (no quotes) as the body of the message.
Received on Thu Sep 10 19:59:19 2009