Re: [computer-go] Pattern Matcher

[Vincent Diepeveen <diep@xxxxxxxxx>]

If i understand correctly you are trying to just predict good moves,
instead of avoiding bad moves?

So in short your program loses every game?

At 18:04 7-11-2004 +0100, Frank de Groot wrote:
>----- Original Message ----- 
From: "Vincent Diepeveen" <diep@xxxxxxxxxxxxxxxxx>
>Subject: Re: [computer-go] Pattern Matcher
>
>
>> My chessprogram just won the dutch open champs and a few months before
>that
>> became 3d in the world champs 2004. Prospects are that in future it will
>> become world champion.
>
>Congratulations.
>Have you noticed you are posting on a Go mailing list?
>
>> How strong is your go program?
>
>How strong is your Go program?
>Don't compare apples with pears.
>The strongest Go program is very bad.
>
>
>> The problem of automatic patterns is the poor *quality* of the patterns.
>
>
>Perhaps when you make an automatic pattern harvester.
>Not when I make one.
>You have religious convictions on my pattern matcher whilst I have the hard
>data to back it up.
>The highest pro-prediction values in the business.
>
>Jealous?
>
>> If the quality of the patterns is 30 kyu, you will never be able to beat a
>> professional dan player.
>
>If my grandmother would have had a dick, it would have been my grandfather.
>Why is my system able to predict most moves in "Mr. Popo vs. GoMonster" when
>trhe quality of the patterns is 30 kyu, hm, wiseguy?
>
>That you managed to make a strong chess program means nothing.
>Give me a break. Chess is a simple game. Any punk with a PC can make a
>strong chess program.
>
>
>> So far no one succeeded in making an automatic system just capable of
>that.
>
>I have made such a system. Telling people they are liars on your religious
>conviction that nobody can be as clever as you are is just plain silly. I
>notice that only the Dutch are as rude as that, so far, here.
>So far you have proven to be an obnoxious punk with exactly zero knowledge
>about patterns in Go.
>
>
>> I hope you realize it's far easier to recognize persons faces, than it is
>> to recognize patterns.
>
>As a matter of fact I have done a lot of work on facial recognition.
>Not the actual recognition of who the face belongs to, but very quickly
>isolating a face from any envinonment, also in worst-case scenario's.
>
>
>> Recognizing go patterns from go games is far harder than face recognition,
>> because you have no clue what you are searching for.
>
>Religious conviction.
>I simply look for everything and keep what occurs most.
>My results prove I was successful.
>
>Can't handle it?
>
>
>> The top 5 of chessworld currently clearly is : Shredder, Diep, Junior,
>> Hydra and Fritz.
>
>Congratulations.
>I couldn't care less what those pathetic little stupid programs do.
>Chess is not Go and Go is not chess.
>So you made a strong chess program.
>Don't be too proud of it.
>Compared to Go, chess is like Tic-Tac-Toe.
>I am not impressed.
>
>Show me your strong Go program when you're so clever, telling others that
>their stuff stinks :)
>If you want to play with the big bous in comp. Go, show us your results.
>A big mouth buys you no credibility.
>
>
>> Automatic pattern learners such as the entire Dan Thies project, they
>never
>> came further than say 1600 rated chess engines.
>
>Chess is not Go.
>Chess is Tic-Tac-Toe. Nothing. Trivial compared to Go.
>
>
>> Wait a minute. Your patterns are automatic, so worse than hand coded
>> patterns for sure.
>
>Nonsenseical religious statement.
>Of course automatic patterns are much better than hand-coded.
>Can you hand-code 8 million patterns and estimate how much each is worth?
>
>In Go?
>Better than my system?
>Don't make yourself ridiculous.
>Stick to chess.
>
>
>> And yes every part in the space shuttle was designed by hand too.
>
>Nonsense.
>The airodynamic parts and the parts that endure stress are designed by
>simulators that DICTATE how parts should look based on Finite Element
>Analysis. So every part that matters is designed by computer. DICTATED how
>it should be by software-driven analysis.
>Your universe is not much bigger than your chess engine.
>
>
>> Some very good patterns were set negative instead of positive.
>
>So?
>Because some chess programmers did something you didnæt agree with, my
>system does not work?
>My results prove it does work.
>
>_______________________________________________
>computer-go mailing list
>computer-go@xxxxxxxxxxxxxxxxx
>http://www.computer-go.org/mailman/listinfo/computer-go/
>
>
_______________________________________________
computer-go mailing list
computer-go@xxxxxxxxxxxxxxxxx
http://www.computer-go.org/mailman/listinfo/computer-go/