It is currently Tue Apr 13, 2021 6:14 pm

All times are UTC - 8 hours [ DST ]




Post new topic Reply to topic  [ 32 posts ]  Go to page 1, 2  Next
Author Message
Offline
 Post subject: Measuring intuition
Post #1 Posted: Sat Oct 10, 2020 4:51 am 
Oza

Posts: 2872
Liked others: 16
Was liked: 3969
I read a lot of game commentaries. I very rarely play. The result is that I have a lot of knowledge about the game but relatively little intuition.

This occurred to me again as a was reading a commentary this morning and thinking, "Yes, I know that", "That's pretty obvious" etc etc. It is in fact rather rare for me to come across a piece of commentary that teaches me something new, and when I read what I regard as the "obvious" stuff, I don't even really have to think about it to see that it obvious.

But if I take a commented game and skip the comments, but try to apply my own thoughts to a position, I feel bewildered and I end up choosing some aspect that is not exactly wrong, but is the not the one the pro chose to comment on.

When I've thought about this problem in the past, I've put it down to not having a good hierarchical structure for the information in my head. I choose the less important over the more important. I don't believe hierarchy can be learned properly consciously. I think this is one of the things that comes with intuition, and since I don't play much I have to accept my handicap.

To give a couple of concrete examples. I will always look at the thickness in a position. This is partly because I have written about it so much. But I have very rarely written about thinness, and so very rarely think about it for myself when looking at games. Yet thinness is one the things pros mention most often. I have said elsewhere that I was surprised at the dominance of certain things in pro commentaries which I mention in my Go Wisdom compilations, which index all the occurrences of each concept in my big books such as Genjo-Chitoku and Games of Shuei. Unlike me they talk of probes a lot. They talk of tsume (checks) a lot. Unlike all amateurs, they tend to talk less of invasions. And so on.

The new thought that occurred to me just now, is that if each of us amateurs were to look at a game and make a list of the concepts that we think of first (i.e. our intuition) at each move, we would be revealing (and ultimately measuring) what goes on in our intuition. Interesting in its own right. But it would then be doubly interesting if we were able to compare each other's lists.

I suspect what we would find is that each band of grades would share common characteristics, and that is why pros can so easily categorise amateurs as of a certain grade after seeing just a few moves.

Anyone game to give it a try? This is one case where the lurkers can't use the excuse of being too weak to speak up.

Top
 Profile  
 
Offline
 Post subject: Re: Measuring intuition
Post #2 Posted: Sat Oct 10, 2020 8:51 am 
Judan

Posts: 6681
Location: Cambridge, UK
Liked others: 432
Was liked: 3662
Rank: UK 4 dan
KGS: Uberdude 4d
OGS: Uberdude 7d
In my commenting whilst playing series of youtube videos I'm basically thinking out loud so you can see the concepts on my mind. https://www.youtube.com/playlist?list=P ... 4AjopUqgbC.

Top
 Profile  
 
Offline
 Post subject: Re: Measuring intuition
Post #3 Posted: Sat Oct 10, 2020 5:17 pm 
Honinbo

Posts: 9216
Liked others: 1550
Was liked: 1596
KGS: Kirby
Tygem: 커비라고해
When you attend a lecture on something (e.g. in school), it's common to think, "yeah, I know that". But then when you quiz yourself later, you can find gaps in what you really know. I think it's the same situation when comparing games with and without commentary. When you have commentary, it's easy to think, "yeah, I know that". But quizzing yourself by asking about the position without commentary is different.

_________________
be immersed

Top
 Profile  
 
Offline
 Post subject: Re: Measuring intuition
Post #4 Posted: Sun Oct 11, 2020 1:36 am 
Lives in sente
User avatar

Posts: 1221
Liked others: 96
Was liked: 401
I did the following exercise: take a game (attachment __go4go_20201010_Yang-Dingxin_Lian-Xiao.sgf, or see below:



Before each move, I tried to come with several possible moves and wrote a comment each time the pro played a move I didn't think at all about. Of course, most of the time the pro's move is better than mine (I didn't mention moves where I briefly thought about the pro's move but preferred another one which is less good). On a few occasions my moves are equally good or better according to Katago (which doesn't mean I fully understand the followups of both options) but that's not the point. In the comments below (see also the attached file Yang-Dingxin_Lian-Xiao - Notes.sgf) I try to understand what is lacking in my thinking so that the pro's move wasn't one of my options. See below (I only looked at moves 1-50 so far).

The comments are my kyu-level interpretations, so are not necessarily correct.



Attachments:
Yang-Dingxin_Lian-Xiao - Notes.sgf [4.27 KiB]
Downloaded 50 times
__go4go_20201010_Yang-Dingxin_Lian-Xiao.sgf [1.19 KiB]
Downloaded 56 times
Top
 Profile  
 
Offline
 Post subject: Re: Measuring intuition
Post #5 Posted: Sun Oct 11, 2020 2:04 am 
Oza

Posts: 2872
Liked others: 16
Was liked: 3969
jlt: Thank you. I wouldn't have got all the pro moves either, but again as soon as I see them I know what's going on.

I once had a pro explain to me some of what they see when amateurs play. One thing was that amateurs tend to be far too aggressive (not just with invasions, but everywhere) and (an important rider) that makes it hard for the pro to teach the amateur in a teaching game because the amateur is trying to wrest control at every stage. The pro needs to have some control in order to lead the games into paths where the amateur can discover new things. The pro still wins, with or without control, because their moves are much more efficient.

Another closely related thing I was told is to remember that go is a two-person game. It's like a conversation. You have to assume the other guy's got something to say. But it's not just that you let the other guy speak. You have to listen to what he says. I think that applies to the moves you overlooked early on on the lower side.

The main characteristic about both bits of advice there is that they are about psychology and not technique (and certainly not about learning josekis!).

As it happens, I've been reading a lot of old Chinese stuff recently (unlike the Japanese, they wrote a lot about the game even in ancient times), and what I noticed, for the first time really, was that they put most emphasis on psychological advice. For example, one piece of advice was not to show happiness when you are winning or grief when you are losing. At first I assumed that was advice to behave with dignity like a Confucian scholar, but when I read on it turned out to be advice not to give the opponent any information on your moves that he might be able to use in the next game!

There is, however, a lot of technical advice, and much of it very different (at least in outward expression) from the Japanese, and it seems to me to much better aligned with AI (though do bear in mind a lot of differences ae to do with group tax). For example, in your move 32 where katago said A was better than the pro move, I think Huang Longshi's adage fitted nicely: if you want to take the centre play on the sides; if you want to take the sides, play in the centre.

If nothing else, the old Chinese theory texts show me there is an awful lot left to learn about go in purely human terms, without AI.

Top
 Profile  
 
Offline
 Post subject: Re: Measuring intuition
Post #6 Posted: Sun Oct 11, 2020 3:37 am 
Dies in gote

Posts: 47
Liked others: 94
Was liked: 7
What is 'obvious' changes over time. At one point it was astoundingly 'obvious' that white opening a handicap game with a 3-3 invasion was 'clearly bad', and yet it's not so clear.

Top
 Profile  
 
Offline
 Post subject: Re: Measuring intuition
Post #7 Posted: Sun Oct 11, 2020 4:47 am 
Oza

Posts: 2872
Liked others: 16
Was liked: 3969
jlt again: What I was talking about was essentially what the Japanese call katte-yomi, i.e. reading for your won convenience - you assume the opponent will play where you want him to play. It may be a characteristic of most amateurs, but straight after that post I came across a pro-level example.



White had just made the triangled forcing moves and was clearly expecting play to continue Black A, White B, in which case White would be well set up to attack Black to left and right.

But Black played C as a probe and to leave aji in the corner. Then he launched the tesuji at D and got the initiative. Interestingly Leela approved and bluelighted this tesuji.

Mind you, the Black player was no less than Huang Longshi.

His opponent was no weakie, either. He had just previously made the early force at E to tempt Black into F now rather than G. It's slightly double-edged in that Black F is stronger against the surrounding White stones, but if White settles himself there, Black will have essentially connected on a dame point - in the fuseki!


This post by John Fairbairn was liked by 2 people: jlt, SoDesuNe
Top
 Profile  
 
Offline
 Post subject: Re: Measuring intuition
Post #8 Posted: Mon Oct 12, 2020 10:02 am 
Gosei

Posts: 1634
Location: Earth
Liked others: 609
Was liked: 279
There is no concept if I play a move intuitive. I just feel where to play. No thinking at all.

Concepts enter my mind when I evaluate my intuitive move candidate. But than it is no longer pure intuition.

Top
 Profile  
 
Offline
 Post subject: Re: Measuring intuition
Post #9 Posted: Mon Oct 12, 2020 10:29 am 
Lives in sente
User avatar

Posts: 1221
Liked others: 96
Was liked: 401
You can't have a candidate move without analyzing, consciously or not, the board position. Anyway the question of this thread is the following : even after careful analysis of the board, my candidate moves were A, B, C, D. The pro played E. Once E is played, I immediately see that E is better than A, B, C, D, which means that if I were able to select E as a candidate move, then I would be much stronger. So why did I fail, what are my blindspots? John mentioned

1) Not trying to understand the opponent's intention.

2) Believing that the opponent will play where I want him to play.

I find these ideas interesting.

Top
 Profile  
 
Offline
 Post subject: Re: Measuring intuition
Post #10 Posted: Mon Oct 12, 2020 10:38 am 
Gosei

Posts: 1634
Location: Earth
Liked others: 609
Was liked: 279
jlt is your intuition for context really that bad? Or are you misunderstanding my statement intentionally?

;-)



And to start the fire even more:

intuition != analysis

Top
 Profile  
 
Offline
 Post subject: Re: Measuring intuition
Post #11 Posted: Mon Oct 12, 2020 10:43 am 
Lives in sente
User avatar

Posts: 1221
Liked others: 96
Was liked: 401
I do have an intuition of what you mean, but after analysis I thought I would add some nuance to your statement.

Top
 Profile  
 
Offline
 Post subject: Re: Measuring intuition
Post #12 Posted: Mon Oct 12, 2020 10:48 am 
Gosei

Posts: 1634
Location: Earth
Liked others: 609
Was liked: 279
Your analysis is potentially misguiding other readers :-) Trust your intuition. ;-)

Top
 Profile  
 
Offline
 Post subject: Re: Measuring intuition
Post #13 Posted: Mon Oct 12, 2020 10:52 am 
Gosei

Posts: 1634
Location: Earth
Liked others: 609
Was liked: 279
But I am genuinly interested in the difference between intuition and rational thinking. And I really felt this thread is lacking a sense for the fundamental dissimilarity between these two modes of operation of our brain.


The concepts that we think of first are not our go intuition.

The moves we think of first are our go intuition.

The concepts we think of first are our intuition for go concepts.

Top
 Profile  
 
Offline
 Post subject: Re: Measuring intuition
Post #14 Posted: Mon Oct 12, 2020 11:12 am 
Lives in sente
User avatar

Posts: 1221
Liked others: 96
Was liked: 401
I think both are inter-dependent. Intuition is shape recognition. For instance:
  • Seeing cutting points
  • Seeing that a group is eyeless
  • Seeing that a group may lack liberties
  • Seeing that a group can surround another
  • Seeing that a shape is cuttable
  • Seeing vital points.

On the other hand, you can improve shape recognition with conscious thinking. For instance
  • Practicing tsumego helps you localize vital points
  • Practicing tesuji helps you see that shapes are cuttable, or that cutting points may be problematic.
  • Learning concepts help you pay more attention to some information on the board. An elementary example: a novice player knows that the winner of a game is the one who makes the most points, and tends to look only at the number of points to evaluate a position. A more experimented player has learned to value influence and strength of groups.

Top
 Profile  
 
Offline
 Post subject: Re: Measuring intuition
Post #15 Posted: Mon Oct 12, 2020 11:22 am 
Gosei

Posts: 1634
Location: Earth
Liked others: 609
Was liked: 279
To measure go intuition I would propose this simple experiment:

Present go positions to players with a very small time frame for their response.



(Or just compare blitz games with slower games of the same players, if you do not want the hazzle)

Top
 Profile  
 
Offline
 Post subject: Re: Measuring intuition
Post #16 Posted: Mon Oct 12, 2020 11:30 am 
Lives in sente
User avatar

Posts: 1221
Liked others: 96
Was liked: 401
Someone in my club says he sometimes plays blitz games, and analyzes them slowly afterwards to find out what kind of bad shapes he made. In short, blitz games reveal what is lacking in your intuition.

I've never applied his method though. Blitz games are too frustrating.

Top
 Profile  
 
Offline
 Post subject: Re: Measuring intuition
Post #17 Posted: Mon Oct 12, 2020 12:06 pm 
Gosei

Posts: 1634
Location: Earth
Liked others: 609
Was liked: 279
A blitz game is won by the player with the better intuition. :-)

Top
 Profile  
 
Offline
 Post subject: Re: Measuring intuition
Post #18 Posted: Mon Oct 12, 2020 12:09 pm 
Lives in gote

Posts: 394
Location: Netherlands
Liked others: 194
Was liked: 106
Rank: EGF 3d
Universal go server handle: gennan
I would say analysis is a more conscious process, while intuition is less conscious. But the boundary between intuition and analysis is not very sharp. And over time, conscious analysis can become less conscious by experience, like muscle memory that a musician acquires as he practices.
Many aspects of a position will be noticed intuitively by a pro (or AI), but requires conscious analysis by a DDK.

Top
 Profile  
 
Offline
 Post subject: Re: Measuring intuition
Post #19 Posted: Mon Oct 12, 2020 12:14 pm 
Lives in sente
User avatar

Posts: 1221
Liked others: 96
Was liked: 401
Gomoto wrote:
A blitz game is won by the player with the better intuition.


More precisely, players with a high winrate at blitz games have a good intuition, but for a given blitz game, luck plays an important role. For instance if a tsumego appears, you don't have time to read so if you have two candidate moves, you have to select a random one.

Top
 Profile  
 
Offline
 Post subject: Re: Measuring intuition
Post #20 Posted: Mon Oct 12, 2020 12:31 pm 
Gosei
User avatar

Posts: 1833
Location: Ghent, Belgium
Liked others: 281
Was liked: 834
Rank: KGS 2d OGS 1d Fox 4d
KGS: Artevelde
OGS: Knotwilg
Online playing schedule: UTC 18:00 - 22:00
Intuition vs rational thinking is very well treated in Kahneman's "Thinking fast & slow", the book I'd save if my house were on fire.

Kahneman's stories, experiments and scientific findings are more about generic thinking and apply to daily life (which makes it such a great read).

Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 32 posts ]  Go to page 1, 2  Next

All times are UTC - 8 hours [ DST ]


Who is online

Users browsing this forum: No registered users and 1 guest


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group