Author: Christophe Theron
Date: 10:16:28 07/05/03
Go up one level in this thread
On July 05, 2003 at 04:32:25, Alessandro Damiani wrote:
>On July 05, 2003 at 00:47:25, Christophe Theron wrote:
>
>>On July 04, 2003 at 15:01:17, Rolf Tueschen wrote:
>>
>>>On July 04, 2003 at 13:35:40, Christophe Theron wrote:
>>>
>>>>On July 04, 2003 at 04:13:13, Rolf Tueschen wrote:
>>>>
>>>>>On July 03, 2003 at 13:04:25, Christophe Theron wrote:
>>>>>
>>>>>>On July 03, 2003 at 05:10:13, Rolf Tueschen wrote:
>>>>>>
>>>>>>>On July 03, 2003 at 01:54:32, Christophe Theron wrote:
>>>>>>>
>>>>>>>>On July 02, 2003 at 04:17:28, Rolf Tueschen wrote:
>>>>>>>>
>>>>>>>>>On July 01, 2003 at 18:20:12, Fernando Alonso wrote:
>>>>>>>>>
>>>>>>>>>>On July 01, 2003 at 15:46:06, Ralph Stoesser wrote:
>>>>>>>>>>
>>>>>>>>>>>Right, that's what I intend with my question. During a 'normal' chess game a
>>>>>>>>>>>chess engine has to face often positions where the difference in evaluation
>>>>>>>>>>>between let's say the 5 best moves or so is very small. In such circumstances a
>>>>>>>>>>>trained neural network maybe could help to find good positional moves better
>>>>>>>>>>>than a classical evaluation.
>>>>>>>>>>>
>>>>>>>>>>>Ralph
>>>>>>>>>>
>>>>>>>>>>I agree with you, that is the important point. To put it in other words, can I (
>>>>>>>>>>a patzer) with my little chessknowledge, beat Fritz 8 using Fritz 8 to analyze
>>>>>>>>>>the moves I my brain "thinks"?. I am sure there is a level of playing were
>>>>>>>>>>someone using a program can beat easily the same program playing alone. But can
>>>>>>>>>>that knowledge be implemented in neural networks?
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>What knowledge? For the moment nobody addressed Tom's objection IMO. Chess is
>>>>>>>>>very concrete.
>>>>>>>>>
>>>>>>>>>Now what you all are saying that there existed a "knowledge" to find "good"
>>>>>>>>>positional moves. Of course our human GM have that knowledge. It is a mixture
>>>>>>>>>out of the evaluation of the very concrete position, deeper (later) consequences
>>>>>>>>>and again very concrete calculations for these _later_ positions. I dont see why
>>>>>>>>>"fuzzy" approaches should do that job better than the "classical" evaluation.
>>>>>>>>>
>>>>>>>>>What you in special are proposing is NOT a question of "knowledge" but simply
>>>>>>>>>one of cheating. You know exactly the "thought process" of a program. So you can
>>>>>>>>>always discover a difference in the evaluation of the final position. Now the
>>>>>>>>>trick is to invite the machine to go blindly for a big difference which is then
>>>>>>>>>the win for you. This is typically the approach of smart amateurs with weaker
>>>>>>>>>chess talents. [Dreihirn comes to mind.] But real chess is something else. A GM
>>>>>>>>>does NOT win because he's a clairvoyant but because his judgement (combining the
>>>>>>>>>very concrete with the general experience for the actual and then later
>>>>>>>>>positions) is "better". A weaker chessplayer has no adaequate judgement at all.
>>>>>>>>>I cant see why neural networks should have one - where should it come from? Out
>>>>>>>>>of the blue?
>>>>>>>>>
>>>>>>>>>Again, you simply didn't address Tom's objection that "sometimes" it is very
>>>>>>>>>important where your Rook is standing. Very concrete. How to handle that
>>>>>>>>>"sometimes" it is "important"?
>>>>>>>>>
>>>>>>>>>Rolf
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>You are demonstrating an almost total ignorance of the subject you are
>>>>>>>>discussing.
>>>>>>>>
>>>>>>>>You'd better read about neural networks first, and only then allow yourself to
>>>>>>>>make comments.
>>>>>>>>
>>>>>>>>But I admit that your writing style is pleasant. Should be enough to convince
>>>>>>>>people who do not know more than you on the subject. Go ahead.
>>>>>>>
>>>>>>>
>>>>>>>You should prefer making statements about topics itself before you utter
>>>>>>>unpleasent statements against other members.
>>>>>>
>>>>>>
>>>>>>
>>>>>>I have made on topic statements by informing the readers that you don't
>>>>>>understand the topic you are talking about and that people should not rely on
>>>>>>your point of view on this particular topic.
>>>>>>
>>>>>>Yes my message was unpleasant, just as unpleasant probably it can be to read
>>>>>>nonsense expressed with authority.
>>>>>
>>>>>
>>>>>Thanks for the bully, dear Christophe. I've waited almost a day to exclude all
>>>>>exaggeration out of my possible reply, but now time has come to teach you
>>>>>certain things you've missed in your education. I'll teach you because you've
>>>>>almost asked for. You need it!
>>>>>
>>>>>Look, Christophe, when I met you in CSS during your questioning for the group I
>>>>>read a bit about you, earlier I didn't know you, and I was astonished. When
>>>>>other people and experts in CC present their vita they give all kind of personal
>>>>>records, but here you explained that your _father_ had founded a specific
>>>>>university down there "in" Africa. While you yourself, correct me if I'm wrong,
>>>>>did not even finish your studies with the usual examinations. Of course that was
>>>>>a very interesting psychological manouevering. Then I "met" you a second time
>>>>>this winter when I was in holidays and bought myself the new 'Science et Vie'
>>>>>and to my astonishment I read in an article about computerchess (with you as the
>>>>>author) mainly something about Chess Tiger, as if Tiger ever had won the Wch and
>>>>>NOT Junior or Fritz or Shredder, of course. That was again a very interesting
>>>>>psychological manouevering. You presented to the French speaking readers a
>>>>>totally twisted history of present computerchess without even mentioning the
>>>>>actual Wch.
>>>>>I give these two examples as a forword so that you can understand what I will
>>>>>now explain.
>>>>>
>>>>>Christophe, if you'd ever been a real (examined) scientist (in whatever field),
>>>>>you would have understood what my commentaries should have meant.
>>>>>
>>>>>Of course I am 1) not a GM chessplayer myself, so no high class chess expert,
>>>>>but I am 2) also no expert for neural networks, so you are quite right with your
>>>>>assumptions. But, dear Christophe, your conclusions are false!
>>>>>
>>>>>I am an examined scientist. I am a psychologist. I am an expert for "debates"
>>>>>between scientists of different fields so to speak. Now all what I did, I agree
>>>>>with you it wasn't very much, I interfered and made clear that the actual
>>>>>answers did NOT yet meet the objection made by Tom Kerrigan. That was the core
>>>>>of my own intervention. Nothing special indeed. Nothing where I had ever
>>>>>expected that a veritable CC expert and programmer could be trapped into.
>>>>>
>>>>>Now you made clear that your comment was "unpleasent". Objection, Christophe! It
>>>>>wasn't unpleasent but stupid. You then say that you made your comment because
>>>>>you wanted that the readers here do know that I am no expert for neural
>>>>>networks. Goodness gracious me! Who had told you this fairy tale that I ever was
>>>>>such an expert? In fact I never was!
>>>>>
>>>>>But you are obviously incapable of understanding why I could still make my
>>>>>comment and thus give a helpful hint into the debate. Hint: my comment did NOT
>>>>>talk about neural networks but about the _debate_ between people about chess and
>>>>>neural network. Can you understand the difference? Obviously not yet!
>>>>>
>>>>>So I can inform you about a new chapter in your education. Today you've learned
>>>>>something about interdisciplinary cooperation. Someone from psychology could
>>>>>participate in a debate between experts of computer sciences, programmers of
>>>>>chess programs and to me unknown members of CCC.
>>>>>
>>>>>This is NOTHING you should get excited about. It's simply possible. It just
>>>>>happens.
>>>>>
>>>>>I wish you good weather down there in Africa or near-by and the necessary fun
>>>>>for your work on computerchess. :)
>>>>>
>>>>>Rolf Tueschen
>>>>
>>>>
>>>>
>>>>I don't understand what my personal background has to do with neural networks.
>>>
>>>hehe. not with neural networks, but your aggressive style of posting with the
>>>judgement that I did not understand neural networks, which I did never pretend
>>>BTW. So I accuse you of producing hot air. And then I show how you do it in many
>>>other respects. Let's see in detail.
>>>
>>>
>>>
>>>
>>>>
>>>>There is one relationship I can see though: on this also you do not know what
>>>>you are talking about.
>>>>* My father has never founded any university.
>>>
>>>But you claimed it in CSS forum!
>>>
>>>
>>>>* I have finished my studies and obtained the degree I was applying for.
>>>
>>>In your staement for CSS you said the contrary.
>>>
>>>
>>>>* I am not the author of the Science & Vie article.
>>>
>>>
>>>I will check this because I am not at home but then I will give a statement.
>>>
>>>
>>>> I just answered a question
>>>>about chess and intelligence and it has been published as a single physical
>>>>column of text. The rest of the article is not mine.
>>>
>>>Objection! You want to say that it wasnt YOU who are responsible for the only
>>>mentioning of TIGER as a chess program???
>>>
>>>>
>>>>But who cares about my personal background here???
>>>
>>>Those who are astonished that you spring on the shoulders of a psychologist who
>>>you declare as incompetent in neural networks although nobody pretended that he
>>>was an expert. The topic was neural networks, but here the question was NOT
>>>about neural networks but about chess, that was no longer addressed and which
>>>therefore was the reason for my message BTW! Not because I am an expert in
>>>neural networks.
>>>
>>>
>>>>
>>>>That you must be an expert for debates, I have no dount about it. That you have
>>>>a talent for creating polemic by bringing up misinformed statements about
>>>>someone else's background, I have no doubt.
>>>
>>>And you yourself, what is your talent besides chess programming? See above.
>>>
>>>
>>>>
>>>>Actually I don't care if you are an expert in psychology, insect sexuality or
>>>>micromolecular astronomy.
>>>>
>>>>You jumped in a discussion about neural networks and made a number of statements
>>>>that show that you don't understand the subject you are talking about.
>>>
>>>
>>>Wrong. I jumped in a topic that was a psychological one, not about neural
>>>networks. Buit if you insist then tell us please what I wrote about neural
>>>networks and what was false. Please give the evidence.
>>>
>>>BTW I dont intend to become n expert in neural networks by just reading 5
>>>articles in the internet. I never heard of such a qualification method! You are
>>>astonishing me.
>>>
>>>Rolf
>>>
>>>
>>>
>>>
>>>>
>>>>Instead of wasting your time answering this message with even more misinformed
>>>>facts, just invest in half an hour of good reading about neural networks. Please
>>>>go to http://www.google.com, type "neural networks basics", click on the
>>>>"Search" button and look at th results.
>>>>
>>>>You might find this
>>>> http://www.generation5.org/content/2000/nnintro.asp
>>>>or this
>>>> http://www.zsolutions.com/neural.htm
>>>>or even this
>>>> http://www.pmsi.fr/neurin2a.htm
>>>>of interest.
>>>>
>>>>But there are hundreds or thousands of pages about the topic of neural networks.
>>>>
>>>>After reading this you will probably understand why your previous statements
>>>>about neural networks are wrong.
>>>>
>>>>
>>>>
>>>> Christophe
>>
>>
>>
>>Once again I'm extremely surprised by how much you insist in digging into my
>>personal background, and by what you want to do with it.
>>
>>You have used 3 wrong facts and armed with this wrong sparse information allow
>>yourself to talk about my "psychological manouevering" and that I presented a
>>"totally twisted history".
>>
>>One thing is now clear: you are as much an expert in neural networks as you are
>>in psychology.
>>
>>As for science, I don't think a real scientist would write anything without at
>>least checking his sources. Or try to conclude anything from three insignificant
>>bits of information (and unfortunately they were not even true).
>>
>>I could explain you why your previous posts showed an almost absolute ignorance
>>about neural networks, but for that I would have to explain how they work.
>>
>>I think it will be much better for you to just read a few articles about the
>>basics of artificial neural networks. By doing so the evidence will pop up in
>>your mind like ad windows do in Internet Explorer.
>>
>>I promiss you won't need to read a lot. Just the basics.
>>
>>I'm sorry. As I said before I think you cannot avoid the trouble of informing
>>yourself a little bit about the topic you were so brilliantly discussing two or
>>three days ago.
>>
>>Unless you are more interested in my irrelevant personal background than in
>>neural networks... But that would be a real loss for neural networks,
>>mathematics, computer chess, philosophy and psychology at large.
>>
>>
>>
>> Christophe
>
>Take it easy, Christophe. Rolf is just more interested in talking than in neural
>networks or computer chess....and through his talking you can look into his ego,
>into his soul. I prefer to look at my code.
>
>Regards,
>Alessandro
I think you are right Alessandro.
Christophe
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.