Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: How to Measure [knowledge based vs bean counter

Author: Amir Ban

Date: 09:42:09 11/18/00

Go up one level in this thread


On November 18, 2000 at 02:54:04, Graham Laight wrote:

>On November 17, 2000 at 19:12:08, Amir Ban wrote:
>
>>On November 17, 2000 at 18:08:46, Graham Laight wrote:
>>
>>>On November 17, 2000 at 13:02:04, Amir Ban wrote:
>>>
>>>>You are missing the point that there do not really exist two separate
>>>>approaches. The term "bean counter" (as used in this thread, it originally meant
>>>>something else) is derogratory. The term "knowledge based" is mostly hype.
>>>>
>>>>Amir
>>>
>>>There are many things that we classify with fuzzy boundaries - but classify them
>>>we do.
>>>
>>>If the term "bean counter" is derogatory, we can use something else (e.g.
>>>"fast").
>>>
>>>"Fast" could be classified in terms of depth, NPS and the like.
>>>
>>>"Knowledgable" could be classified in terms of the number of discrete pieces of
>>>knowledge a program uses (which I think is a better measure than lines of code).
>>>
>>
>>Notice that you are now assuming that "knowledge" and "fast" are opposites. How
>>do you know this ? If there's anything I said in this thread, it's that this is
>>not true.
>
>Actually, I didn't mean to imply that.
>
>There's probably more correlation between speed and knowledge than there is
>between, say, board colour and playing strength - but you're certainly not going
>to get a correlation of 1.
>
>>The notion of gauging knowledge by size of evaluation function or pieces of
>>knowledge is even sillier than gauging strength by node count.
>
>I don't agree. In the case of Crafty's evaluate.c program, because it is written
>so clearly, I think one can count the discrete pieces of knowledge with
>confidence.
>

I meant that the count is meaningless regardless of how clear its value is.

Amir


>-g
>
>>Amir
>>
>>
>>>People often dislike being classified with fuzzy boundaries - but if so, they
>>>would be hippocritical if they then classified other people in this way ("old
>>>man", "young man", "boy", "genius", "idiot" etc).
>>>
>>>-g



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.