Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: SSDF. Gandalfs LEARNING is destoyed when aborting!!

Author: Mogens Larsen

Date: 08:48:59 10/18/00

Go up one level in this thread


On October 18, 2000 at 10:01:58, Thorsten Czub wrote:

>I DO INTERPRET ?
>
>mogens: do you have a brain ? i guess so.
>now : do you see the world outside ? i guess so.
>can it be that you INTERPRET the world outside with your brain ?
>could be , or ?
>
>now: ANYTHING is interpretation. there is nothing OBJECTIVE that cannot
>be seen without subjective interpretation of human brains.
>
>show me a thing that is NOT interpreted by human brains.
>if you find something let me know. than i would call this a new information.

Of course everything is interpretation, but there are correct interpretations
and there are wrong interpretations. In this case your interpretation
contradicts the factual basis on which the data is created. To be precise. You
can't invent a interpretation when you know it contradicts the abilities of the
subject itself. Human beings can't fly, jumping from a tall building
interpreting your arms as wings won't help.

The principal variation of a knowledge based program isn't a plan. It's a
calculation like the one performed by "bean counters". Only difference is the
addition or altering of specific variables and parameters. Nothing more and
nothing less. The moves might appear more sensible and intelligent, because the
choice of move is based on more information. That doesn't mean planning at all,
just a more educated guess. The programmer stipulate preferred values by adding
parameters. The program just calculates accordingly.

In fact you could argue that longterm plans demands accuracy, not speculative
evaluation. Speculative evaluation could be defined as the destroyer of "plans".
Does this mean that "bean counters" are constructing plans through accuracy? No,
it doesn't. There are no specific goals or intentions to reached by planning,
only calculation.

>:-))) paradigms get changed themselves. the old dinosaurs die out,
>and new paradigm come automatically. you cannot stop this.

I have no such intentions. I prefer intelligent programs myself, but the
diversity makes it more interesting to watch. Both speculative and accurate
evaluation has its merits. Ignoring one or the other would be stupid.

>I think it fits pretty well for me to like knowledge-based and
>intelligent-chess programs, because i don`t like bean-counter programs that play
>silly.

You're probably right. Those without knowledge seek it. Happy hunting THORsten.

Regards,
Mogens



This page took 0 seconds to execute

Last modified: Thu, 15 Apr 21 08:11:13 -0700

Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.