Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: A Practical, Implementable Self Learning Chess Computer

Author: Ricardo Gibert

Date: 10:48:36 10/18/00

Go up one level in this thread


On October 17, 2000 at 09:22:30, Vincent Diepeveen wrote:

>On October 16, 2000 at 06:12:20, Graham Laight wrote:
>
>>Following discussions in the "recognising chess patterns" thread last week, I
>>believe we have reached the stage where it possible, with today's PCs, to create
>
>I'm sorry, based upon what conclusion can computers get smart?
>
>I mean you can learn till infinite times, but still do nothing useful
>with the knowledge: how do you plan to INTERPRET the knowledge?
>
>I mean suppose you have a few tera of knowledge, how do you DISCRIMINATE
>between what is good and what is wrong?
>
>Where is your decision algorithm?
>
>This is the whole point. I've never seen a good decision algorithm
>to decide which pattern is good and what pattern is wrong.
>
>>a self learning chess computer. Before, I have always had doubts about whether
>>it could be done with current technology - now these doubts are gone.
>
>I think a machine which only knows true of false will never get smart
>like we humans are. Nothing compares to human mind! That i think i
>concluded not because i'm religious, but basically based upon tens of
>thousands of tries. Something is wrong or something is true on the
>computer. Nothing between that.
>
>Working with percentiles like the neural networks can do also sucks bigtime,
>as it still is not doing any statements between wrong and good.
>
>So on the one hand we want a clear decision to be taken, but we don't
>want to do that based upon heuristical bounds!
>
>>Disappointingly, the discussion thread died out - possibly because people didn't
>>realise what was in the thread.
>>
>>So here's the deal. I will think through the outline design of such a system,
>>and write it up in my (hopefully!) clear style for everyone's perusal if, and
>>ONLY if, at least 5 people promise to comment on the design after I have written
>>it.
>
>i'm always in for betatesting learning systems as long as they can
>run under either windoze or linux. my email: diep@xs4all.nl
>
>>So, if you want to read it, and you're willing to comment on what I write,
>>please indicate this by responding to this message.
>
>here you have the reply of someone who's prepared to test, but at the
>same time thinks you want to build a scyscraper from water in the
>sahara desert.

This is an entertaining analogy. I'll have to remember this one.

>
>>Regards,
>>Graham



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.