Re: Guessing?

On Jul 14, 9:42 pm, Marshall <marshall.spi..._at_gmail.com> wrote:
> On Jul 14, 10:30 am, JOG <j..._at_cs.nott.ac.uk> wrote:
>
>
>
> > On Jul 14, 5:45 pm, Marshall <marshall.spi..._at_gmail.com> wrote:
> > > On Jul 13, 9:07 am, JOG <j..._at_cs.nott.ac.uk> wrote:
>
> > > > >> The greatest weakness in the entire debate, however,
> > > > >> is the capacity issue. Lack of computing capacity is
> > > > >> a complete explanation for what computers can't do (yet.)
>
> > > > <splutter/>
>
> > > > Ok, this one is just ridiculous. Lets take the bastion of good old
> > > > fashioned AI - chess. In the 90's the chess AI "deep blue" was
> > > > processing over 200 million board positions a second. That's right.
> > > > 200 millions every single second. Let's compare that to a grand
> > > > master, who can examine about 8. Yup, that's 199,999,992 less
> > > > positions per second than the AI.
>
> > > Hey! You've been complaining about the other side's simplistic
> > > analyses, but here you're doing exactly the same thing. Deep
> > > Blue included special purpose hardware for playing chess, as
> > > well as dozens of general purpose CPUs. And you're claiming
> > > it's looking at 25 million times as many positions per second.
> > > Yet, Deep Blue lost to Kasparov, and Deeper Blue only just
> > > managed to eke out a victory. So, the 25 million number is
> > > crap, isn't it?
>
> > C'monnnn, its incredible. Examining 8 positions per second vs 200
> > million.
>
> You did not address my point *at all.*

Not my intention to miss it. I assume therefore that your saying that because Kasparov and Deeper Blue are closely matched that "25 million times faster" can't be true, or must represent low capacity? If so this seems so illogical that I must have just glossed over it (I'll elucidate in a mo).

> Kasparov and Deep Blue
> are roughly at parity in terms of winning.

Yes in results, but not in capacity. In terms of brute force processing of chess positions Deep Blue is light years ahead (yes, I know, still not answered... see down).

> You are making
> exactly the mistake that you complain about in the field:
> invalid comparisons.

Always possible, but I don't think this is likely. What do you think I am comparing incorrectly and I'll try and address it.

>
> We see over and over again that special-purpose hardware
> provides enormous benefits in computational power. A
> factorization engine with custom built hardware can out-factor
> a general purpose computer running factoring software
> by an enormous margin. If you know exactly the problem
> domain ahead of time and can craft a solution directly to it,
> that gives a lot of leverage. Kasparov is not special-purpose
> hardware, and yet he still was at parity in the problem domain.
> Therefore he *obviously* is using higher-capacity hardware.

The logic here seems totally awry to me (Who are you and what have you done with Marshall!). Consider:

• 2 computers, A and B.
• Give both a million numbers and tell them to sort them.
• Computer B takes longer.
• Therefore Computer A is *obviously* running higher-capacity hardware?

And nothing seems off there? Like maybe computer B was using a bubblesort,  and computer A a more efficient strategy? Given the empirical evidence, it seems much more likely that it is the /way/ that kasparov plays chess that keeps him competitive despite his inability to parse as large a search tree. That he has less capacity but a better strategy.

>
> > Waving our hands about and saying "if only we just had more
> > processors" or "if only we could just add /another/ hundred thousand
> > rules", is just wishful thinking (bordering on religious faith). The
> > problem is not capacity, it is approach.
>
> Approach may well be a problem; there are so many different
> approaches being taken that it is unclear to me. However capacity
> is certainly a problem, regardless of approach. And stop using
> the inflammatory word "religious" thank you very much.

Hey, that seems a bit rich given you started with the "magical" and "mystical" stuff back when you were calling people idiots.

>
> > I agree that there is no
> > reason to think AI is not achievable,
>
> Then we're done.

We were done in the post when you conceded you made a mistake and that noone was denying AI wasn't possible. Since then I assumed we were just exploring the ideas about the best ways of getting there.

Regards, Jim.

>
> Marshall
Received on Mon Jul 14 2008 - 23:41:17 CEST