It is currently Thu Mar 28, 2024 5:50 am

All times are UTC - 8 hours [ DST ]




Post new topic Reply to topic  [ 19 posts ] 
Author Message
Offline
 Post subject: Takemiya's experiment
Post #1 Posted: Mon Oct 07, 2019 4:00 am 
Oza

Posts: 3647
Liked others: 20
Was liked: 4626
I have just read an interesting experiment involving Takemiya Masaki, which perhaps indicates how young pros are adapting to AI.

Way back in 1985, the now defunct Igo Club, which used to be my favourite magazine, presented an article which asked where to play an erasing move in the following position. It was from a game by Takemiya Masaki which I either haven't got or yet done for the GoGoD database. I'm too old/lazy to climb into the loft and rummage through lots of boxes of magazines to refer to the original article (the game is apparently from 1985, but the opponent is not specified).



It may have been just a next-move problem then, but a new article now - part of a series coordinated by Hirata Tomoya - asks Takemiya and young pros how far they would dare enter White's moyo as an erasing move. Obviously the focus was quite different in 1985 - go AI wasn't even a gleam in the eye - and the article raises the interesting point whether simple "next-move" questions are always appropriate. There are plenty of positions that suggest more than move and the way to discriminate depends on the flow or the "story" of the game. (The Japanese text uses the English word "story" but "narrative" is probably better English.)

But Hirata now asked several young players where to play with a view to seeing whether there were differences between Showa and Reiwa (i.e. pre-A and AI generation players). As I say, I have declined to hunt up the original, but Takemiya's own Showa sequence was Black A, White B, Black C. We all know that this vague sort of move is the sort that gets highlighted in commentaries, so we may want to infer that Takemiya was more a than a little pleased with it. But that was then - his Showa intuition. His Reiwa intuiton has changed.

Before you read on you may wish to make your own guess, but make sure it's backed by a positional judgement first. Maybe you'd want to choose both a Showa and a Reiwa move :)

There may be a flaw in the methodology in that there is no mention of komi. The 1985 game would have been played with, at most, 5.5 komi. But Hirata spotted, just like Lizzie, that Black is in dire straits and so needs to wrench control of the game in some way. A quiet erasure is not enough, but one that is too deep (an invasion, really) in White's strong area is courting disaster. It seems to be taken for granted that Black's move does have to be inside this moyo, and Lizzie certainly agrees that this is the main area but she also gives close scrutiny to several moves on the left side.



Reiwa Takemiya now prefers the forceful shoulder hit at A, as does young player Fukuoka Kotaro 1-dan (aged 13). Fukuoka's view was that things are easy for Black if White answers on the third line as Black keeps pressing on the fourth line, and if White caps instead (a standard response) Black can achieve shinogi. Ah, the blithe confidence of youth! (He had his eye on F as a key move.)

Takemiya only mentioned White's third-line response, but instead of Fukuoka's forceful subsequent play, he showed only a vague jump out to the centre. He was actually somewhat surprised to be told he'd played as in the first diagram 34 years ago!

Ueno Asami 3-dan got the evaluation right - she'd prefer to be White - and suggested B for Black. Her strategic thinking seemed to be that w White surrounding move on the right side would out the game out of sight for White, so an erasing move tout court was required. But in intuition terms her first thought was the shoulder hit at A. She eschewed that because a White cap would make her too nervous, especially because of the White wall (but which wall? - answer at the end).

Onishi Ryuhei, 4-dan and rising fast (youngest ever Shinjin-O; he's 19), took a rather different strategic tack. He too obviously sensed Black was n trouble and so wanted to cause confusion, so he opted for Black C, expecting then White D, Black E. He felt this gave Black options of messing around on the upper right side while also aiming at the cut at G. But actually, his initial intuition was the same as Asami's - shoulder hit at A but spooked by a White cap.

My version of Lizzie preferred the shoulder hit. It didn't even list Takemiya's Showa move as a candidate. No surprise there. But what might surprise some is that when this Showa move was forced upon Lizzie, she didn't bat an eyelid and gave the same win rate as before!

This latter phenomenon (rating unconsidered moves highly) is too common - troubling even - to ignore. It needs a name so we can talk about it more. I think it was Bill Spight who first noticed it, and is certainly pointing it out most often, so I propose we call it Spight Analysis, or something like Spight Retrospective Analysis. When we use this tool, human pros can usually be shown to be performing very often only a whisker away from AI-bot level.

It also suggests to me that people like Taemiya may have allowed themselves to be swept along by the fashion for AI-type moves. Maybe he ought to have had more confidence in his original judgement?


This post by John Fairbairn was liked by: jeromie
Top
 Profile  
 
Offline
 Post subject: Re: Takemiya's experiment
Post #2 Posted: Mon Oct 07, 2019 5:57 am 
Honinbo

Posts: 10905
Liked others: 3651
Was liked: 3374
John Fairbairn wrote:
I have just read an interesting experiment involving Takemiya Masaki, which perhaps indicates how young pros are adapting to AI.


Interesting comparison of the Ghost of Takemiya Past with the Ghost of Takemiya Present. :)

Quote:
My version of Lizzie preferred the shoulder hit. It didn't even list Takemiya's Showa move as a candidate. No surprise there. But what might surprise some is that when this Showa move was forced upon Lizzie, she didn't bat an eyelid and gave the same win rate as before!

This latter phenomenon (rating unconsidered moves highly) is too common - troubling even - to ignore. It needs a name so we can talk about it more. I think it was Bill Spight who first noticed it, and is certainly pointing it out most often, so I propose we call it Spight Analysis, or something like Spight Retrospective Analysis. When we use this tool, human pros can usually be shown to be performing very often only a whisker away from AI-bot level.


Thanks for the vote, John, but I think that the phenomenon is well known in chess, where computers have ruled for some time. Still, it is something I expected because of training bots by self play. Obviously, that has been good enough to produce superhuman play and to improve current programs, but for analysis I would like a program that can discover such overlooked alternatives, and perhaps even improvements on the bot's first choice. OC, a program that searches too broadly would be too slow to analyze a whole game in a reasonable time. But maybe a human analyst could have the option of having the program do a broad search for selected positions. :)

_________________
The Adkins Principle:
At some point, doesn't thinking have to go on?
— Winona Adkins

Visualize whirled peas.

Everything with love. Stay safe.

Top
 Profile  
 
Offline
 Post subject: Re: Takemiya's experiment
Post #3 Posted: Mon Oct 07, 2019 6:27 am 
Judan

Posts: 6725
Location: Cambridge, UK
Liked others: 436
Was liked: 3719
Rank: UK 4 dan
KGS: Uberdude 4d
OGS: Uberdude 7d
Is there a white stone missing at L16? (Black's shape is weird otherwise, and white only has 17 stones on the board to black's 19, assume 1 white captured top right). Also when you say Lizzie liked Takemiya's old move as much as its suggestions I think you need some qualification of network and playouts as I got quite different results: I just checked with LZ #226 with a few hundred playouts and it thought about 10% worse than shoulder hit. And that black was ahead (missing L16?)

P.S there's a review mode modification of the LZ engine that searches more widely.


This post by Uberdude was liked by: Bill Spight
Top
 Profile  
 
Offline
 Post subject: Re: Takemiya's experiment
Post #4 Posted: Mon Oct 07, 2019 9:30 am 
Oza

Posts: 3647
Liked others: 20
Was liked: 4626
Quote:
Is there a white stone missing at L16?


You are right. And White has also had one stone captured. Sorry.

Top
 Profile  
 
Offline
 Post subject: Re: Takemiya's experiment
Post #5 Posted: Mon Oct 07, 2019 3:55 pm 
Oza
User avatar

Posts: 2408
Location: Ghent, Belgium
Liked others: 359
Was liked: 1019
Rank: KGS 2d OGS 1d Fox 4d
KGS: Artevelde
OGS: Knotwilg
Online playing schedule: UTC 18:00 - 22:00
This is my no-Lizzie but yes-cgoban analysis


Top
 Profile  
 
Offline
 Post subject: Re: Takemiya's experiment
Post #6 Posted: Tue Oct 08, 2019 10:57 am 
Lives in sente

Posts: 905
Liked others: 22
Was liked: 168
Rank: panda 5 dan
IGS: kvasir
Not sure what this position has to do with AI. The shoulder hit is the most direct approach, maybe ideal if you are more worried about playing slack than making a mistake. Takemiya's original move looks super confident, maybe he was playing a weaker opponent? The rest of the game is not included so we can only guess what the plan was or the result.

Leela on my mobile phone still wants action on the left side after Takemiya's moves. Not sure what to make of that.

John Fairbairn wrote:
This latter phenomenon (rating unconsidered moves highly) is too common - troubling even - to ignore. It needs a name so we can talk about it more.

I think that is called search bias, the AI has a bias to consider some moves and not others. We can call it blind spots when a good move is not found or the percentage changes dramatically when the variation is played out. It seems negative to talk about blind spots if the AI can find some other good move, just not the one we want it to find, so I would just call it bias in that case.


This post by kvasir was liked by: Knotwilg
Top
 Profile  
 
Offline
 Post subject: Re: Takemiya's experiment
Post #7 Posted: Tue Oct 08, 2019 12:12 pm 
Gosei

Posts: 1733
Location: Earth
Liked others: 621
Was liked: 310
(KataGo Winrates attached to Knotwilgs file)


This post by Gomoto was liked by: Bill Spight
Top
 Profile  
 
Offline
 Post subject: Re: Takemiya's experiment
Post #8 Posted: Mon Oct 14, 2019 6:08 am 
Oza

Posts: 3647
Liked others: 20
Was liked: 4626
Here's another example from the series:



Takemiya played A in 1960 (Showa; again an unknown game - presumably this was to avoid the other players knowing the original move) but in Reiwa would now play B. His "feeling" was essentially that the Reiwa choice changed the direction of play somewhat. Perhaps this is more of a centre-facing "fighting shape" as recommended by Murashima (see Shuei and AI thread).

The male teenyboppers Fukuoka Kotaro and Onishi Ryuhei both chose C but for quite different reasons and envisaged quite different follow-ups. Fukuoka saw it as the "positive" way to play. Onishi, who correctly (if we rely on Lizzie) saw that Black was already a little behind, was influenced by the fact that some shapes had already been settled and so he reasoned that the upper side would be split 50-50. That in turn would make Black strong in the upper right corner, and he wanted to take advantage of that by mapping out a moyo on the right side. For him, C was a stake in his fence rather than an attack on White. Fukuoka in contrast had his eyes on an invasion at the lower 3-3 point.

Ueno Asami needed to spend just 5 seconds to decide on D - "the only move" - simply because allowing a White approach up there was unconscionable.

Hirata Tomoya had a similar mindset about not allowing the opponent to play first in his chosen area, but chose the lower right (like Takemiya) with E.

All these moves except Takemiya's were Lizzie picks (and she added some others on the lower left side), but using "Spight analysis" (looking at the winrate after playing Takemiya's moves anyway) just confirmed it was a confusing position for both her and the humans. Each move was within a very narrow band, well within the margin of error Bill has talked about. Furthermore, Lizzie's first choice for a long time was C and it was only after a deep search that she switched to E (54.5% over C's 53.9%).

My cynical conclusion is that, at least in this sort of position, it all seems to be a matter of style. Otherwise we have to conclude that pros don't just not know why bots choose their moves. They don't even know why other pros choose their moves!


This post by John Fairbairn was liked by 2 people: Bill Spight, ez4u
Top
 Profile  
 
Offline
 Post subject: Re: Takemiya's experiment
Post #9 Posted: Mon Oct 14, 2019 6:22 am 
Honinbo

Posts: 10905
Liked others: 3651
Was liked: 3374
Gomoto wrote:
(KataGo Winrates attached to Knotwilgs file)


Thanks. :)

Do all of these numbers come from analyzing a single position, or are some moves in the variations input by a human and a new analysis made from there?

_________________
The Adkins Principle:
At some point, doesn't thinking have to go on?
— Winona Adkins

Visualize whirled peas.

Everything with love. Stay safe.

Top
 Profile  
 
Offline
 Post subject: Re: Takemiya's experiment
Post #10 Posted: Mon Oct 14, 2019 6:55 am 
Oza
User avatar

Posts: 2401
Location: Tokyo, Japan
Liked others: 2338
Was liked: 1332
Rank: Jp 6 dan
KGS: ez4u
John Fairbairn wrote:
Here's another example from the series:

...
Takemiya played A in 1960 (Showa; again an unknown game - presumably this was to avoid the other players knowing the original move) but in Reiwa would now play B....

Takemiya was 9 years old in 1960. Is the date incorrect?

_________________
Dave Sigaty
"Short-lived are both the praiser and the praised, and rememberer and the remembered..."
- Marcus Aurelius; Meditations, VIII 21

Top
 Profile  
 
Offline
 Post subject: Re: Takemiya's experiment
Post #11 Posted: Mon Oct 14, 2019 6:56 am 
Gosei

Posts: 1733
Location: Earth
Liked others: 621
Was liked: 310
Quote:
Do all of these numbers come from analyzing a single position, or are some moves in the variations input by a human and a new analysis made from there?


I load the sgf file into Lizzie and walk through all variations. I did not enter any moves additional to the file.

(If I would like to explore a move more deeply with more playouts I would enter the move by hand to force KataGo to have a deeper look. But for quick first impressions I just browse the file.)


This post by Gomoto was liked by: Bill Spight
Top
 Profile  
 
Offline
 Post subject: Re: Takemiya's experiment
Post #12 Posted: Mon Oct 14, 2019 11:10 am 
Oza

Posts: 3647
Liked others: 20
Was liked: 4626
Quote:
Takemiya was 9 years old in 1960. Is the date incorrect


My booboo. It was Showa 60 - 1985.

Top
 Profile  
 
Offline
 Post subject: Re: Takemiya's experiment
Post #13 Posted: Mon Oct 14, 2019 3:02 pm 
Lives in gote

Posts: 502
Liked others: 1
Was liked: 153
Rank: KGS 2k
GD Posts: 100
KGS: Tryss
John Fairbairn wrote:
All these moves except Takemiya's were Lizzie picks (and she added some others on the lower left side), but using "Spight analysis" (looking at the winrate after playing Takemiya's moves anyway) just confirmed it was a confusing position for both her and the humans. Each move was within a very narrow band, well within the margin of error Bill has talked about. Furthermore, Lizzie's first choice for a long time was C and it was only after a deep search that she switched to E (54.5% over C's 53.9%).

My cynical conclusion is that, at least in this sort of position, it all seems to be a matter of style. Otherwise we have to conclude that pros don't just not know why bots choose their moves. They don't even know why other pros choose their moves!


I wouldn't call this a "confusing position", but rather an ambiguous position : several choices are possible.

Katago analysis with 30k playouts :

Attachment:
Katago-Takemiya.png
Katago-Takemiya.png [ 900.2 KiB | Viewed 9278 times ]


And for exhaustivity :

- A has 716 visits, wr 50.2% and 0.8pts
- D has 130 visits, wr 50.2% and 0.8pts

Top
 Profile  
 
Offline
 Post subject: Re: Takemiya's experiment
Post #14 Posted: Mon Oct 14, 2019 5:54 pm 
Oza
User avatar

Posts: 2401
Location: Tokyo, Japan
Liked others: 2338
Was liked: 1332
Rank: Jp 6 dan
KGS: ez4u
Tryss wrote:
...
I wouldn't call this a "confusing position", but rather an ambiguous position : several choices are possible.

Katago analysis with 30k playouts :

...

I think you left the komi at the default 7.5 points. When I change the komi to 5.5 points, I get an interesting result. Katago thinks Black is ahead by 4 to 5 points with a winrate around 60%. In it's searching, it solidly favors M4 (above E and right of B). It never invests much time on A or B (1K+ in each). At the time I snapped this picture, it had about 7K visits in C and 4K in D. ALL of these points are evaluated between 60.1% and 60.7%, so flip a coin!
Attachment:
Takemiya 2 2019-10-15_9-39-34.jpg
Takemiya 2 2019-10-15_9-39-34.jpg [ 365.49 KiB | Viewed 9261 times ]


Edit:
I forgot and left Lizzie running after writing this post. After about 225K visits, katago switched to slightly preferring Ueno's D! If you look closely, you can see that it estimates the score at +4.6 points for M4 at the bottom vs +4.4 points for D at the top. However, on winrate it gives the tip to D over M4.
Attachment:
Takemiya 2b 2019-10-15_10-23-44.jpg
Takemiya 2b 2019-10-15_10-23-44.jpg [ 366.08 KiB | Viewed 9257 times ]

_________________
Dave Sigaty
"Short-lived are both the praiser and the praised, and rememberer and the remembered..."
- Marcus Aurelius; Meditations, VIII 21

Top
 Profile  
 
Offline
 Post subject: Re: Takemiya's experiment
Post #15 Posted: Mon Oct 14, 2019 8:49 pm 
Honinbo

Posts: 10905
Liked others: 3651
Was liked: 3374
ez4u wrote:
Edit:
I forgot and left Lizzie running after writing this post.


Maybe that's the key! :mrgreen:

_________________
The Adkins Principle:
At some point, doesn't thinking have to go on?
— Winona Adkins

Visualize whirled peas.

Everything with love. Stay safe.


This post by Bill Spight was liked by: Gomoto
Top
 Profile  
 
Offline
 Post subject: Re: Takemiya's experiment
Post #16 Posted: Tue Oct 22, 2019 5:54 am 
Oza

Posts: 3647
Liked others: 20
Was liked: 4626
There's been a further instalment in the series and this interested me particularly for a special reason.



It was White to play, and in 1985 Takemiya chose A. Nowadays he would prefer B. LZ (which puts Black a smidgeon ahead) says that loses about three percentage points. All the young players also chose A, but Fukuoka Kotaro actually vacillated between that and B, and he did mention come concern about the left side.

LZ, however, had a full percentage point preference for C, which was the move chosen by Hirata Tomoya 7-dan.

It seems that most of the human pros either had a fixation with the visually dominating wall, or underestimated the aji of the cut at D, or just felt too confused by the left side to be able to pick a move. LZ of course has no psychological hang-ups.

The reason this interested me specially is an emerging finding from my work on the Go Wisdom appendixes I am now adding to books of pro commentaries I do. I index every mention of a term that appears in a commentary and discuss the term, usually in great detail, to help readers of the book think about the games with all that extra information at their disposal. But there is fascinating information simply in the index alone. Sheer numbers of occurrences tell us something, and insights can also be gleaned from how early or how late terms occur (e.g. probes on move 4, boundary plays on move 14, fuseki play on move 83, etc).

Although I haven't yet done a proper analysis, one thing that that keeps striking me over and over again is that the three commonest types of mistakes made by pros are as follows:

Three types seem to stand out.

1. Probes. Strong pros commenting on weak pros often chastise them for not making probes.

2. Order of moves. This is usually of the type that can be analysed by tewari. Wrong timing of forcing moves or other intransitive changes is much less common, and less criticised.

3. Direction of play. And this is how I view the example above.

Analysing why pros so often get the DOP wrong (as adjudged by fellow humans) might be a monumental task, but I'm willing to hazard a guess that psychological factors of the kind mentioned above come into it. I think it is also fair to say that their mistakes are normally of a very small order of magnitude, and certainly way below the level in amateur play. But mistakes are mistakes and I'm sure the pros would also like to know why they make them. My early impression is that they haven't found AI much help yet in that regard.

It's going off at a bit of tangent, but here's an "easy" (10-kyu to dan level) direction of play problem by Rin Kanketsu.


)
This is interesting for various reasons. One is that the White invasion on the lower side is scorned by LZ which prefers either of the 3-3 invasions on the right side - less "direction of play" than "different planet!". But maybe Rin chose this position as something weak amateurs would encounter.

LZ initially agrees with Rin that A is the right. B is plain wrong and doesn't even appear on LZ's radar (although E does in that area).

C is rejected as inferior by Rin (and LZ seems to agree) because forcing contact plays are better to hinder sabaki by White.

But the biggest surprise may be that, after a deeper search, LZ's preference (by a full percentage point) switches to C. I find that hard to talk about in DOP terms. But maybe it's a probe, or even an order of moves issue! In the variation LZ shows, Black still gets a wall across the 4th/5th lines, but gets the left corner as well, whereas in the human commentary Black merely encroaches on the corner.

Another reason I gave this extra problem is that it features sanrensei by Black. I have seen comments here that seem to suggest it is now regarded as bad. I obviously can't say whether it is or it isn't, but FWIW pro comments I have read seem to judge as too difficult rather than wrong (like tengen), though LZ seems to think it loses a fraction over one percentage point (yet is that not within the margin of error Bill has talked about, especially so early in the game; and it may be stylistic?) But I am more intrigued by the fact that pros are still playing it.


This post by John Fairbairn was liked by 2 people: Bill Spight, Gomoto
Top
 Profile  
 
Offline
 Post subject: Re: Takemiya's experiment
Post #17 Posted: Tue Oct 22, 2019 7:08 am 
Lives in gote

Posts: 502
Liked others: 1
Was liked: 153
Rank: KGS 2k
GD Posts: 100
KGS: Tryss
In this position, with 90k playouts, Katago prefer D, with 66k visits, 43% winrate and -2 pts. Followed by E (19k visits, 42.5% wr, -2.2pts).

Katago don't like C, it's barely explored, and when played, the evaluation don't change much (even with 20k playouts). Same with A, even if it get 0.5% better than the initial evaluation with 40k playouts.

Top
 Profile  
 
Offline
 Post subject: Re: Takemiya's experiment
Post #18 Posted: Tue Oct 22, 2019 8:08 am 
Oza
User avatar

Posts: 2408
Location: Ghent, Belgium
Liked others: 359
Was liked: 1019
Rank: KGS 2d OGS 1d Fox 4d
KGS: Artevelde
OGS: Knotwilg
Online playing schedule: UTC 18:00 - 22:00
A thought.

A major difference between humans and bots may be that we may assign a higher value to something that preserves the past as we are loss averse, while the bots only care about the future prospects and have no notion of a particular loss.

In this case, we see White's "invasion", which makes us aware of the potential loss of something that was temporarily Black's. Now I know the Japanese "invasion" is more about dislodging stones than stealing territory, so the Japanese/pro treatment may be closer to the bots' than "our" Western narrow view on invasions. Be what may, for low dans like me, moves A, C and even B, carry the notion of "preserving what's ours" in their evaluation.

The bots are known to think "corners, sides, centre" so they will naturally evaluate D and E first or higher at first, and evaluate A or C for their merits of strengthening the black stones and weakening White's. From this perspective a pure side territorial move like B is understandably omitted from the analysis.

This may explain why K4 is not highly rated to start with: when White invades, K4 becomes a stone to strengthen. Going for D directly instead will strengthen the approach already, establishing a position in the corner.

Top
 Profile  
 
Offline
 Post subject: Re: Takemiya's experiment
Post #19 Posted: Tue Oct 22, 2019 9:41 am 
Honinbo

Posts: 10905
Liked others: 3651
Was liked: 3374
John Fairbairn wrote:
The reason this interested me specially is an emerging finding from my work on the Go Wisdom appendixes I am now adding to books of pro commentaries I do. I index every mention of a term that appears in a commentary and discuss the term, usually in great detail, to help readers of the book think about the games with all that extra information at their disposal. But there is fascinating information simply in the index alone. Sheer numbers of occurrences tell us something, and insights can also be gleaned from how early or how late terms occur (e.g. probes on move 4, boundary plays on move 14, fuseki play on move 83, etc).


Interesting and valuable work, John. :)

Quote:
Although I haven't yet done a proper analysis, one thing that that keeps striking me over and over again is that the three commonest types of mistakes made by pros are as follows:

Three types seem to stand out.

1. Probes. Strong pros commenting on weak pros often chastise them for not making probes.

2. Order of moves. This is usually of the type that can be analysed by tewari. Wrong timing of forcing moves or other intransitive changes is much less common, and less criticised.

3. Direction of play. And this is how I view the example above.

Analysing why pros so often get the DOP wrong (as adjudged by fellow humans) might be a monumental task, but I'm willing to hazard a guess that psychological factors of the kind mentioned above come into it. I think it is also fair to say that their mistakes are normally of a very small order of magnitude, and certainly way below the level in amateur play.


From the Elf commentaries and the AlphaGo games I think that direction of play is one of those concepts that will undergo quite a revision in coming years. By and large, bots don't seem to care about it, or care very much. And sometimes they violate our understanding of it. Often if I have to guess which side a bot will block a 3-3 invasion on, I will now guess that it's the one that does not conform to direction of play (as I understand it). ;) But the preference is usually quite small, so in those cases they don't really care much at all. I think that this is related to the bots' devaluation of the sides. If the sides don't matter much to the choice of play in one corner, how much do the adjacent corners, twice as far away?

Also, in line with Knotwilg's comment, I suspect that a number of pro mistakes stem from the human psychological factor of loss aversion. Bots are much more willing to throw matters up in the air, confident that they will come down better than they are now. Humans like to hang on to what we've got. Amateurs are obviously more prone to this failing than pros, but I suspect, from what obsevation I have done, that pros are more prone to it than bots.

Quote:
But mistakes are mistakes and I'm sure the pros would also like to know why they make them. My early impression is that they haven't found AI much help yet in that regard.


Taking a look at pro games from last year, I have been impressed at how few fuseki mistakes they now make, according to Elf. That's really quite rapid progress. :)

One major problem, OC, is that AI choices do not come with language, i.e., with well thought out concepts. Our first reaction is to try to fit them within our current concepts and to modify our current concepts to fit the bots' plays. Children, OTOH, can just pick things up by imitation. A decade from now it will be interesting to hear what pros who grew up with AI have to say. Exciting times!

_________________
The Adkins Principle:
At some point, doesn't thinking have to go on?
— Winona Adkins

Visualize whirled peas.

Everything with love. Stay safe.

Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 19 posts ] 

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:  
cron
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group