"Pattern recognition" DEBUNKED

Sort:
urk
Ziryab, just because you played the correct move doesn't mean you needed to have seen Horvath also play the correct move in a similar situation.

You were able to find the move more quickly because you recognized it, granted, but it doesn't change the fact that it was the one and only best move. Pattern recognition only helps, but it doesn't explain good moves versus bad moves, at all.

I don't see how your Horvath example means anything.
fieldsofforce
fieldsofforce wrote:
fieldsofforce wrote:

In the movie Rainman, in the scene where the toothpicks  were scattered  on the diner floor, the  savant , Raymond (Dustin Hoffman) blurted out 82, 82, 82, 246.  The waitress  confirmed that there were only  4 toothpicks left in a box that originally contained 250.  The savant's brain is wired to group the toothpicks in patterns of 3 groups of 82 each.  Our average brains just see a jumble of toothpicks scattered on the floor.

 

 

urk
Fieldsofforce, and how many idiot savants, with their vaunted PATTERN RECOGNITION, can play chess at a GM level? The answer is NONE. They cannot understand the game.

Thanks for helping me make my case.
fieldsofforce

Thanks for ignoring the facts.  You are entitled to an opinion, but NOT your own facts.

urk
The human with the best pattern recognition in the world (maybe Rainman?) can't play chess worth a crap, no matter how much he tries, guaranteed.

You want me to play a match against Rainman? I will.
Sqod

I didn't read the article, but I can already tell there's a huge false assumption there: the human subconscious is many times more powerful than the conscious, and most of what we learn is subconscious, so attempting to explicitly list the patterns that were learned, and/or to count them, then to make conclusions on the results is ridiculous. The subconscious is presumably how humans acquire commonsense knowledge, which is one of the primary hurdles of artificial intelligence: trying to get a machine to have commonsense reasoning. 

http://www.zdnet.com/article/the-next-frontier-for-artificial-intelligence-learning-humans-common-sense/

Ziryab

urk, you're not paying attention. No one says that pattern recognition explains "good moves versus bad moves." Rather, the discipline of looking at many positions--in whole games, tactics exercises, opening study, and endgame training--helps you see more quickly your candidate moves. Often, calculation is still required, but often enough no calculation is necessary because the pattern is simple (due to study) and forcing.

 

 

In your original post, you reference an article that deserves wide readership. The author makes a reasonable case using one particular book that definitions of patterns are fuzzy. In the end, however, he offers "structures" as a better term, which he promises to elucidate in a future article. In logic, this argument is called bait and switch. Focusing the argument on one particular book that appears to have been written somewhat hastily is called a straw man.

 

urk
Ziryab, you haven't said anything that refutes my contention that pattern recognition does not explain GM level play.

You used the word "discipline" and I think this goes to explain why so many people so easily overrate pattern recognition. As a teacher you appreciate good study practices. Sure, study improves your game, every GM studies, but they win because they understand better how the pieces interact, not because of memorized pattern recognition.

Open your eyes.
tiredofjapan

Ultimately, it's a semantic argument.  I think there's merit to an argument of "soft pattern recognition", which would posit that the ability to see repeating factors of position is one tool which may be of great importance in playing chess.  If you set up an argument of "hard pattern recognition" in which memorized positions and actions are of great importance, then I would prefer to agree with the statement that pattern recognition is of dubious value.  It seems to me that what most participants in this thread want to agree with is the soft pattern recognition argument, but the fault line is set up as hard pattern recognition or no pattern recognition.  I would posit both sides as a straw man, arriving at a synthesis in which each captures an element of the truth of the matter.

Ziryab

"One of the most important things in chess is pattern recognition: the ability to recognise typical themes and images on the board, characteristics of a position and their consequences."
https://en.chessbase.com/post/magnus-carlsen-on-his-che-career

urk
In the past I've gone along and agreed when somebody uttered the magic words "pattern recognition."
How stupid of me.
fieldsofforce
urk wrote:
In the past I've gone along and agreed when somebody uttered the magic words "pattern recognition."
How stupid of me.

Being a contrarian and denying a well established concept in  chess is not  against the law.

Also, being an asshole is not against the law.

Candidate35

My analogy works exactly like that of chess but the anology   just made it painfully clear what you are suggesting compared to reality. You're seemingly being so obtuse about the matter that you refuse to either acknowledge the points being made to you or are so blinded by your apparent "discovery" that all you see as correct is your view. Can you cite a GM that hasn't studied many games, or looked at lots of chess puzzles in their lives? I've never known of any myself, but I've known a multitude who have.

 

You can't be a good chess player and NOT know patterns in chess. Many patterns, at that. But keep arguing otherwise if you wish. I'll let others carry on this discussion if they wish for I've lost interest.

 

Ziryab

Another common pattern in these threads is an occasional brilliant retort. In this case micky1943 nailed it.

kindaspongey

Perhaps it would be helpful to look at Improve Your Chess Pattern Recognition by Arthur Van de Oudeweetering
https://www.newinchess.com/Shop/Images/Pdfs/9006.pdf
and Train Your Chess Pattern Recognition by Arthur Van de Oudeweetering.
https://www.newinchess.com/Shop/Images/Pdfs/9021.pdf

razzarainbow

I used to visualise horses wining races and wthin three months the horse would win exactly like my dream .My ex wife visîted me one day and the horse was at the post I told her my dream of wat would occur .In my dream the horse finished 3rd but also in my dream I was paid as a winner. The horse did in fact finish 3rd but te 1st two strayed from the hurdlès course and were disqualified

razzarainbow

My prophetic dreams weren't commonplace maybe a bout of 3 every 2 yrs then went to 1 every 10 years ,I.m 64 ,now zilch .nada nothing atall mores te pity .

razzarainbow

my point in relation to this thread is maybe its not really our talent to begin with maybe some form of 'possession from an deceased family member or a benign spirit ,if thats not too weird

SAGM001

I dream a lot of chess Positions . I play chess in my dreams sometimes and when I wake up , I can set up the Positions sometimes too that I was playing in the dream . :/

razzarainbow

Prince and Paul Mc Cartney both dreamt songs,Mc Cartney only one Yesterday, Knowing that we can jump time has no logical explaination and hearing music from out of nowhere is bizarre