I think "initiative" is a good thing to have in go, but what does it mean, exactly?

Let's consider a 9-stone game between a 15-kyu player and a pro. It's, let's say, 20 moves into the game. The board position is still good for black, but the pro has been making strides toward closing the gap. Who has the initiative?

I would imagine that, at least in some cases, it's the pro who has the initiative here. With the wording used here, the pro has "control of where the game is headed" - if not from the board position, at least from the dynamic flow of the game.

However, if we are to analyze the board position at that particular point in time, the bot evaluation may very well be "B+7" or "B+8".

I'd personally consider white to have the "initiative" here, but for the bot evaluation to still favor black. They are different measurements.

That being said, I don't have a precise definition for "initiative", so some may not consider white to have the initiative in the hypothetical example that I am giving here. My current feeling for the word might be somewhat related to "momentum".

_________________
be immersed

I've forgotten where I first read this but I treasure it as some of the best Go advice I've ever received

I don't think you are "wrong" per-se but maybe the conclusion is too strong or the premise too simple.

The error of an estimate is not only due to the variance of the estimator but also the bias of the estimator. Bias of an estimator is the difference between the average estimate and the true value. You account correctly for the case when the estimator is unbiased but very likely this complex estimator that is Go AI is biased in ways that depends on the position (and other factors such as the number of playouts).

If for example we start in position of type A where the estimator is biased against moyos and then the game evolves into position of type B where the bias is close to zero, then it could be expected that this favors one player (the one playing with the moyo) in terms of how the estimates change and maybe in terms of actual game results. An example of type of position with bias that is close to zero would be the endgame! It might be optimistic to think this can be detected easily or that the effect is significant but I remember we discussed this in a thread about old videos on how to become a dan player

Another way that the estimate can drift is when the estimate in a position of type A has high variance but the game evolves into position of type B that has low variance. In this case the estimate in the first position will likely have a larger error than in the second position, if the stars align (which might not be that unlikely) then this can look like one player was favored and one could draw incorrect conclusions (about mistakes in the game or that the AI estimate favors something that it actually does not).


=Edit I wrote it favored the player playing against the moyo but on reflection it is the other way around and then that probably also favors that player in terms of game results. So I just changed it, hoping the point comes across.

embrace the numbers
and
praise the emergent joy of go

We have to understand what we're talking about . . .

In both chess and go, there is actually no such thing as winning probability. A game is either 100% winning for black, 100% winning for white, or a draw. This means means that any such concept of a winning probability is purely a concoction of the human brain or AI processors inability to know the perfect sequence of play. Winning probabilities are entirely a personal thing, to either the AI or the computer, and it doesn't make sense to nail yourself to another person or computers winning probabilities, regardless of strength, because each person has different learned styles. And even the style you've currently learned to play should be double checked, because much of if may be based on poor learning that left you playing with a style that is not in line with your natural, innate style based on your phenotype--or rather your genotype, since your phenotype may also be suffering from the same issues as your learned style. Tami is absolutely accurate that you must focus on style first before skill level. skill is style since style limits skill.

Of course, the above conclusion assumes that when we say 'winning' we're referring only to winning probability. In actuality, in Chess and Go, the control you have on the board when you win is half of what's meant. In chess it's square control, and potential square control represented by material, in go it''s territory or maybe area in general.

So if your training an AI to win by the maximum amount of territory it can, the purest way to do so is for it to multiply the probability of a position occurring by the score difference. So a position in which you win by 6 points but have only a a 50% chance of reaching is worth 3 points. If the is a semeai in which to a certain perspective the result is either a 50% chance between the game ends in a draw or one player wins by 50 points, then from that certain perspective that player is ahead by 25 points. But actually, what about the perspective of the player who's winning? Losing? The weaker player, but rather maybe the stronger players perspective is what we should take as who's winning. In essence, if we base who's winning purely on the likelihood to win the game, it's an absolute myriad since that means when an even game begins between a stronger player and a weaker player, by definition the stronger player is already winning and the weaker player losing, and that's not what we mean by a person being in a winning position. The alternative that works is if you assume perfect play and whether someone is winning or losing is in absolute terms, but then does away with the concept of winning or losing by a little or a lot, so that has to be applied on top of this concept, which kind of brings you back full circle. Note also AI winning probabilities aren't necessarily AI winning probabilities.

Let's you were watching the women's continental football tournament. One team has more points on the board, but at the point you're watch, the team with less goals changed their strategy, got motived is actually now playing way better to the point where it's obvious their going to win. So who's winning?

All this is says that if an AI or a pro says don't use so-and-so opening because for games at their level their is a slight disadvantage--you can be assured you needn't lose any sleep if that's what you use. In fact, I'd advise you to use this rule of thumb; anything less than first move advantage, in chess and go, is definitely on the table! So we should spend time stressing over AI's opinions and AI 'cancel? culture' where entire joseki's are wiped out because the AI deems it loses by 0.2 percentage points--okay I'm obviously exaggerating, but still, switch that time to thinking about what it means to win in go. The exact same applies to the social culture as well, ironically both became prevalent after March 2016 so it's interesting.

I'd hypothesize that a similar argument could be made for most expressions of probability, as probability is a measure of uncertainty.

In the real world, if you roll a six-sided die, and have full knowledge of every physical dynamic of the world (gravity, hand friction, effect of the atmosphere, etc.), you would know with 100% certainty the result of the roll.

But that stuff is too complicated to quantify, so we have a simplified model, measuring probability in the absence of all of those minute real world variables.

_________________
be immersed

Yes, I should have phrased it instead as probability is an inherently subjective viewpoint of a physical system. Unless of course one believes in the fundamentality of quantum indeterminism, but determinism is the theory that most follows Occam's razor. From a perceptual sense however outside of mathematics all is probability.

Baduk TV often has the opinion of 4 AI's running simultaneously, which is also how I think it should be used.

Baduk TV often has the opinion of 4 AI's running simultaneously, which is also how I think it should be used. Just treat them like another pro with their own biases, not an oracle. And now I'm mixing up the plurality of my pronouns like someone who just got up from bed, even though I haven't

Why would it be difficult to understand? Ishida Yoshio had a series of articles in Go World called "Bad Moves: How many points do they lose?" where he wasn't shy about analyzing certain moves that were considered bad and putting points values on how much they lost. This is the same thing, except the player is probably stronger than Ishida and you can ask about any position you like.

Well, lightvector seemed to agree it's not straightforward, but in any case I don't see that the comparison with Ishida is really valid. He takes local positions not whole games, admits other pros have different opinions, and uses numbers as a journalistic gimmick, simply in lieu of a table he also kindly provides. Minor bad moves are 2 or 3 point losses, bad moves are 5 point losses and very bad moves are 10-point losses. He does not explain his counts, beyond using the criteria of this table. So apples and pears come to mind.

Furthermore, Ishida claims moves that lose ten points are few and far between (which suggests to me he may have been at the cider ), and his aim is not to teach an understanding of counting, but to make the amateur realise that his main fault is not making bad moves per se, but making an accumulation of minor bad moves and bad moves. This is valuable information that AI is not able to verbalise.

W+5.7

What does it mean?
it depends

John, I'll have to disagree with you on this.

"...He takes local positions not whole games..." AI does this also, and better than humans do

"...admits other pros have different opinions..." Just as different AIs may have different opinions. Though this seems to be less true as AI gets better.

"...He does not explain his counts..." To some extent AI doesn't either, but AI numbers can be traced back to something meaningful.

"...his aim is not to teach an understanding of counting..." This is similar to AI.

"...but to make the amateur realise that his main fault is not making bad moves per se, but making an accumulation of minor bad moves and bad moves." This is something that AI point loss and win rate makes clear at least to me. Some moves are worse than others, but the accumulation of even small errors over the course of a game leads to a loss against a much better opponent.

"This is valuable information that AI is not able to verbalise." While AI won't verbalize it, I think AI is uniquely able to show us that it is true.

Fine, but you don't seem to be disagreeing with me really. My main point was that I didn't think the comparison between AI and Ishida was valid - apples and pears. You haven't mentioned that, so pro tem I'll assume you agree.

I could debate the other points, but instead I'll try to take a different approach that may let a bit more light in. I'll ask you to try to view things from my point of view.

My go activity can be split into three unequal parts. The smallest by far is reading SL and, very occasionally, other internet material. I get very little out of this in go terms, but I have an abiding interest in the way other people think and for that SL does provide material.

I read Japanese go magazines, newspapers and books, and sometimes Chinese material. Nowadays I rarely look at Korean material. This is the bulk of my go activity.

I spend a lot of time writing books. Overall, I probably spend more time on this than anything else, but a high proportion of this time is actually devoted to reading books and magazines. For simplicity, therefore, let us say I spend my time between reading SL and reading Japanese material. I rarely look at AI programs and don't watch videos (can't hear them, but prefer books anyway).

What I see when I contrast these two sources (SL vs Japanese) is a chasm. SL seems dominated by people who have mathematical interests or careers of some kind (I include programming in that) in the real world. Most of these people love AI and spend a lot of time on it, but given the nature of the SL readership (as I see it), they are preaching to the already converted.

When I look at Japanese material, I see very, very little reference to AI or numbers or rules or maths of any kind. None of these elements is run down in any way, and any of them can be respectfully highlighted at times (though in a quite different way from SL). The people who write this Japanese stuff (professional journalists) are preaching to the generality of people.

If I look a little more deeply, I see SL as a group of amateurs who are not themselves improving as a result of AI and who are not helping to grow the game. I see the Japanese press as well supported by professionals (some of whom actually are improving with AI!) who are not really preaching but explaining, and who above all are successfully growing the game.

Now, I have to stress that there is no question in anyone's mind that AI is stronger than any professional, probably by a large margin, by the criterion of winning games. But humans make mistakes for all sorts of extraneous reasons (e.g. time trouble) and so lose games they might win in perfect circumstances. By that measure, they are not so distant from AI as many people make out. That aspect tends to get lost on SL.

There is also no question that AI is fascinating in many ways. But the fascination is strongest for those already inclined that way and who have a relevant background. The fascination for everyone else is much less, can be minimal for many, and actually off-putting for some (personally, I'm somewhere in the middle). That gets lost on SL.

In the professionally produced Japanese material, I think nothing gets lost. It is well balanced. Apart from study material, which seems to be what SL denizens are mainly concerned with, the journalists provide news, history, interviews, crosswords, etc ,but there is also balance within commentaries - the occasional reference to AI evaluation, occasional AI charts and so on. I try quite often to bring some of these other aspects of go to SL (and books), but in the main it gets ignored. My fault, or is it the make-up of the SL readership, or the western readership in general?

Now if we go back to the original title of this thread (Is this opening really so bad?), I see that as a topic that could easily appear in a Japanese magazine. But when I envision how that question would be answered there, I see something that almost belongs to a different universe from SL. On SL, we get rival AI evaluations, decimal points and general disregard for professional expertise. I try to red flag that, but I'm the one who gets red flagged.

But I don't really mind. I've got my Japanese material, so instead of gruel for every meal, I still have what I regard a well-balanced diet.

Does that help you understand how I see things? For me, it's not really an argument between AI and pros. It's about the SL treatment versus the more balanced treatment.

One could say an AI doesn't explain either, at least without effort to experiment with the position. But in truth, Ishida actually does try to explain how he arrives at his counts, with explanations along the lines of "This territory is worth 15 points, but the two stones by themselves would have been worth about 10, so Black 1 here has only gained 5 points. I assess the loss of strengthening White on the outside as 15 points."

I imagine the focus on one side of the board is mostly a question of limited space in print. They are opening/early middlegame problems so the reader seems to be invited to imagine an otherwise mostly empty board. Seems like a reach to call this an important distinction.

It's easily seen when playing the AI and watching your estimated score go down over time, with all kinds of step sizes.

But again, my point was: why overcomplicate this and not just say "it's a strong player's opinion about the position, for whatever that's worth", in either case.

Growing the game would be nice but it is out of scope for a thread where someone simply asks whether an opening is bad. (My own attempts at growing the game include writing a piece of software that includes beginner tutorials, and, pre-COVID, going to my city's annual Japan Festival and teaching Go to beginners). And I wouldn't be able to tell who here is improving or not, and for what reason, so I think it's not worthwhile to engage in such speculation.
Are they? For a long time the general consensus seemed to be that Go in Japan was in decline, and if anything it was Hikaru no Go that made a dent in it. But what you are overlooking is that people are using the tools and resources available to them. Japanese press or Chinese literature is not readable to the vast majority of people here. AI evaluations and game reviews are available easily and essentially for free, unlike pro teaching.
The one exception is Michael Redmond, who makes video content for USGO and his own channel - the man is a treasure and as far as I can tell he seems to be well loved by the community. But he also references AI evaluations so maybe you wouldn't be a fan.
I don't think anyone disregards professionals - it's a matter of accessibility. But most people here do not worship them either, that is true. (Ishida, by the way, also has decimal points, when he says a move loses "perhaps a half or two thirds of a point").

This really belongs in a different thread, but a HUGE amount of JCK material is available in English, in books and Go World. It is still mostly available, but western go players have chosen to put western go publishers out of business.



Yes, but you get what you pay for. Water for most of us is essentially free, but most people prefer champagne, beer, coffee, tea, etc. (There are even people who like Prosecco!!!!!!) There is water in them thar drinks, of course, and I don't object to AI being used similarly for commentaries as an ingredient. But I prefer to pay for extra taste.

This is in line with my perception that the reason we have a plethora of go apps in the west despite the smaller western go population, to the point of where the cutting edge of go AI started in the west and even today Asian pros use western-made go AI, is that the go population in the west is dominated by computer types.

But I adopt the Carlson philosophy that large size is not inherently good. Growth is good only if it's healthy, and paves the way for future growth. However, if people in the west have the perception that only computer, then a large go population of computer types is not as good as a population half the size with a more diverse mix of people, since in the former, there is an absolute limit to the potential of possible people who could 'fit in'. Computer nerds. I find it ironic go players in the west seem to only think taking short-term losses for long-term gains applies only over the go board. I say this with the greatest admiration for them, as I think my brain is allergic to the concept of computer programming, so see computer programmer and espicially teen hackers as divine beings far beyond what I could possibly reach, more impressive that the teen Asian go pros. But still.

What I don't agree with John Fairbairn--well, perhaps what I agree with him more than he does himself--is waiting until AlphaGo to raise this observation more diligently. It's unfair to suddenly expect 'numbers guys' to suddenly change and be who they're not. The solution to the overdependence on 'numbers thinking' is not downing there valuable contribution but rather balancing the 'numbers people' with 'poetry people', 'philosophy people', 'new agey guru people', people who won't be too hung up on whether AI thinks an opening is +0.2 points or not, people.

I only have about 1 year 6 months left so I'm going to start being more candid about my history.

When I had just learned go at around 12 years of age, I use to pay a lot more attention to women's, go for two embarrassing reasons. One is being able to identify as a demographic that is traditionally not as strong at go as a go player in Europe, but more majorly is that children identify their future selves with men and their current selves with women, for obvious biological reasons.

In this case, there should be a concerted effort to promote go among people in artistic or scientific fields that are underrepresented in. The people who claim to be awake all the time and never sleep—they think they're Nasir Jones—focus on genetic diversity, but I'm more into diversity of thought in this particular case, despite ethnology being my main interest. However fixing the go delusion is the west is more important to me now in the time I have left.

I'm currently less intelligent than I was when I was a child, but apart from that haven't really changed much. But a few perceptions have changed. I think that that majority of activity in the west to improve western go is a complete waste of time and is even discriminatory. I would spend all resources on only:

1-People who are neuro non-atypical or with special needs
2-Women and youth go, they can't realistically be separated
3-Apsects of go that don't require you to be a 'go player' to appreciate. For one I hate the wrong and ridiculous assumption that when a beginner learns go, the first thing you should do is try to make them stronger, since being stronger is how you can appreciate the game. These same people will then complain about when they think online go players had misbehaved in some way!?!? That's what what happens when your go culture is dumb enough to obsess over 'getting stronger' to the point where it's assumed for everyone. Or maybe it's a consequence of computer types disregarding poetic and artistic elements of go.

Now I've taken things off topic enough, if we want to evaluate what the computer program actually means when it says +5.7, it could mean 'There are a huge range of possible endings with possible scores including were both white and black wins by huge margins, but if you average it out, the difference if +5.7 for white,' or it could mean 'I'm fairly confident that there are only a few good lines in which white wins by 5 or 6 points', but the latter is normally only during the endgame. By definition of how AI trains, it is the former. This is also probably what Cho chikun means by being able to tell a game is within half a point even though the middlegame has barely ended, the Japanese pros expect us to make moderate mental effort and use common sense if we claim to want to get stronger in a mindsport. This is What Ishida Yoshia means when he gives points to opening patterns, blahdy blahdy blah . . .

I for one do think it would probably be far more valuable to learn how Chinese and Korean pros talk and think about go rather than Japanese pros, for obvious reasons. However we must also diffrentiate and discern how pros talk about go between themselves and how they present their thoughts with a general audience in mind--we shouldn't assume it's the same thing