We've weak evidence (from Demis Hassabis's go teacher) that AlphaGo is around british 3D (maybe 4D KGS) around February, we now know that it's around 9D in October. What we don't know is the AlphaGo growth rate, it maybe constant with time, exponential (scary) or just plateau out in the end. So the March match is very important to keep track of what's going on, and to predict what will come in the future.

But I'll guess this just for fun

Exponential Rate --> unlikely, but if it does happen, more than 14D in March, it's immeasurable using the traditional ranking system
Constant Rate --> around 13D, enough to crush Lee Sedol to 5-0 though
Plateau --> around 11D, the outcome might go either way but I'll bet AlphaGo win at 3-2

Yes, extrapolation is fun - not so much more.

I have just been watching the AGA video with a Korean 9p saying AlphaGo plays with a "Japanese" style. If this is attributable to the training set of games being skewed towards Japanese pros playing online ... then from some points of view there is a potential increase in strength by redoing all that with a more broadly based set of games.

Well, one could see why this would be an obvious move, for a Korean pro!

Looking at it another way, versions of the AI trained on different sets might show fine differences of level. Which could perhaps be calibrated by reference to pro ranks.

So, only a small amount of external consultancy might have suggested some path or paths to retraining, and I don't suppose we can guess the resulting increment. Traditionally, a third of a stone means plenty in terms of pro levels.

Charles

We also don't know if we have any valid data on which to base that "growth rate".

The machine(s) used for training (what the program is running on for training and testing) not necessarily the same throughout that time. We don't that 3-4D British in February and 9D in October was on the same machine.

Go is played with time controls, time management is part of playing go, and machine power trades with time. We mustn't confuse "program" with "program + machine". Google is willing and able to throw money at the problem, and when running in the October contest used MUCH MORE machine than we have seen the other programs using in the periodic computer program vs computer program contests or for the bots playing on KGS. They can afford the "iron", the others can't.

When we say "this PROGRAM is stronger that that PROGRAM" we should really be adding "when both have comparable computer resources available to them".

I do think this new program IS stronger than the existing MCTS progams. But I think the "how much stronger" is being distorted by the hardware disparity. Thus while I will be watching the outcome of the upcoming match with great interest, I would actually rather see a contest between this program and say AYAMC5 both running on say a single i7-4790 machine. Better? How many stones better? << that would be a machine significantly more powerful than most of our home computers, but a serious gamer might have something like that >>

Fan Hui said AlphaGo has a peaceful style - I guess that means the same thing.

I wonder if there is something inherent about a strong AI program that causes it to be more peaceful. Possibilities:

- Perhaps what we view as messy just isn't that messy to AlphaGo, and it simply chooses the most straightforward path to a good outcome.

- Perhaps playing to its strengths means keeping a small lead is good enough, because it is probably unflappable in the endgame.

I think there is a bit of confusion over what the training games for the network was. It seems to me it was KGS 6-9d games, rather than pro games. That strikes me as a bit odd as I'd have thought the higher quality of play in pro games in GoGoD, go4go, some bulk download of Tygem 9d(P) games etc would lead to a stronger AI but perhaps there was not enough quantity. In the AlphaGo paper it says:

Most KGS 6-9d will be amateurs, though perhaps at the 9d level more are pro. But I don't think there are many Japanese players there, in fact most of the known top KGS players are European/American/Chinese (rapyuta was a Japanese insei iirc). It could be that KGS has a more 'Japanese' style than Tygem though.
Myungwan Kim and Andrew Jackson in their review (https://www.youtube.com/watch?v=NHRHUHW6HQE very interesting, great job guys!) seemed to think it was trained on GoGoD pro games, which would indeed probably have more Japanese pro games, but they may be mistaken.

macelee tells us he gave them a go4go dump in December 2015, so it's possible that the version that beat Fan Hui was only trained on the relatively weak KGS 6-9d games and the one that will face Lee Sedol in March will be trained on a higher-quality diet of professional games and be even stronger. Gosh.

I wonder how different the network in AlphaGo is from the one a few of the DeepMind folks described in their earlier paper:
http://www.cs.toronto.edu/~cmaddis/pubs/deepgo.pdf
When we discussed that paper at end 2014 / early 2015 there was some scepticism over what a move prediction accuracy of 55% meant when it came to playing strength. Perhaps back then their policy network plus MCTS was around 3d and then the development of the value network (plus refinements) since then has given it the boost up Fan Hui beating strength.

Most programmers know their own plateau point of their program (where throwing a better hardware doesn't make it stronger), which is why in all computer Go tournament (as I know) they prefer to bring their own hardware. I'm aware that in chess there's a single system competition though.
In their research, they give Zen 8 cores and Crazystone 32 cores, which is fair enough because they can only acquire the commercial version which only scale at that point.
Mogo use to run on up to 3200 cores supercomputer, but it became a solid evidence that MCT is not scale well on high performance computer.
If AlphaGo happen to scale well on supercomputer, then it's a program strength too! And I've no problem with Google throwing their computer farm at it, because they can. However, according to the paper, the alphaGo reach the plateau as well (they give 700 more cpu core and 100 more gpu core but the ELO only increase from 3140 to 3168)

I don't understand all this scalability non-sense. It is clear in the paper, and in the previous page of this topic, that single-machine AlphaGo plays at Fan Hui level, winning the majority of 4 stone handicap games against Zen, etc.

This is not a matter of throwing more machines at the problem.

A Boeing 747 and a Piper Cub are both "single aircraft". But there is a vast difference in their load carrying capability. That's a very big "one machine" compared to what everybody else is using.

By scalable, means a more or less linear change in performance with changes in computation power. Probably neither is very scalable. Nor do I know how close Zen is to the "diminishing returns" point. But I do know for one of the other MCTS programs. Running on an old Lenovo T61, MFOG12 plays at about 1 dan KGS. Give it about 10 times more crunch power like what the bot on KGS has and is is 2-3 dan. However make that another 10 times, maybe would go up only half a stone (that's what I mean by diminishing returns).

I voted for Lee Sedol by 3-2, though I hope he wins 5-0.

This opinion is worth little, since I am 4k at Go and it is 25 years since
I worked with a simple neural net. And as many have observed, we can only
guess at how much the program and its computer resources will improve by March.

There are two reasons I favour the human this time. Perhaps the human
must be good enough to find one or more valid tesuji or joseki innovations
that were not in the program's training set. Lee Sedol may be capable of that,
Fan Hui less so. Even world-level human players make mistakes (in the judgement
of their peers, or find them when reflecting on their games), which is why
the human may need to play more than one unprecedented masterstroke per game
to beat a program which never blunders.

The other reason is that analysis published by the British Go Association
http://www.britgo.org/files/2016/deepmind/BGJ174-AlphaGo.pdf criticises
several of AlphGo's moves, so it is not in fact perfect either. Comments by Younggil An
also identify mistakes by the program in Game 5, while also noting that it played
the endgame perfectly.

On the other hand, Lee Hajin has pointed out that AlphaGo may only have played
as strongly as was necessary to win, and it might automatically play better
against a stronger human.

So, 3-2.

I do hope that the match terms are clear and are clearly followed, so we do
not get the kind of bad taste that was left after Kasparov complained about
his loss to Deep Blue. For example, the terms should allow the human to rest
or sleep, and the computers to be restarted if they crash or the power fails, but
they should not allow the program to be adjusted on the fly by a secret cabal
of experts. (Whether such adjustments actually happened against Kasparov or
would even be possible this time is beside the point.)

I think Lee should go for a very weird opening right from the start. The engine wouldn't be able to deal with it with the current data set. I voted 3-2 for AlphaGo.

I think it's mostly clear that AlphaGo seems to play more conservatively when it thinks it's significantly ahead and perhaps aggressively when behind.

The paper says that it was trained on KGS games, but it used a weighting dependent on the strength of the players. I suggested that they ought to use GoGoD games too, but that was subsequent to the match.

I don't see any reason for Lee to vary from what he would do against another middle-ranking pro. Game 1 will probably be, for him, a question of trying to establish how strong AlphaGo is. The alpha-male strategy of trying to win convincingly at the start of the match is, amusingly, something AlphaGo is immune to.



The most interesting thing, apart from the scoreline, will be to see AlphaGo put through its paces. I like the way it plays: generally lucid and with a stable feel even in heavy fighting.

I'm not yet quite ready to articulate where the style comes from. I suspect it is "thematic", a step up from "shapeful". In other words it may have learned that certain types of things "in the offing" should affect the way you play.

David Ward went round Korean clubs: and he reported to me that the local strong amateurs whom he played adopted a steady style, looking to punish errors, rather than be provocative/disruptive. This is also pro common sense for the "stranger in town" opponent, I suppose.

So, all told, there may not be the playing to the gallery, at least at the start.

One question I haven't yet found an answer to is when Alpha started training against itself, i.e. How much time did it take to achieve it's current level (I haven't read the paper yet).

_________________
On Go proverbs:
"A fine Gotation is a diamond in the hand of a dan of wit and a pebble in the hand of a kyu" —Joseph Raux misquoted.

That observation is less useful than it might appear. Consider the following substitution

I think it's mostly clear that "a strong go player" seems to play more conservatively when "he/she" thinks "he/she is" significantly ahead and perhaps aggressively when behind. We could even replace "go" with "any game where the result is either win or lose and margin considered irrelevant".

In other words, simply the correct way to play. A consequence of AlphaGo being a strong player.

This is admittedly unrelated, but it reminds me of the story of the NHL glow-puck back in the 90s.

I also noticed that Google recently overtook Apple as the most valuable company in the world by market capitalization. I can't help but wonder if part of that rise in share price* was thanks to the DeepMind announcement, which I'm sure went down well with investors. Spending $1m for $billions of stock growth is a great return.

* though most of the overtaking of Apple was due to Apple going down.

Match will be on "9, 10, 12, 13, 15 March - will be livestreamed on YouTube." from https://twitter.com/demishassabis/statu ... 7870008321

It now has 5 wins and 9 losses, and rates him at 3013. If the four wins from the European professional championship get added, who knows how high he could go ?

_________________
Occupy Babel!

What if the match ends in a tie? Will there be a tie-break game?

Reportedly, Chinese rules will be used, which should mean that no long cycles other than sending-two-returning-one are prohibited.

The match involving a computer, it should follow the superko rule. I guess that all repetitions are prohibited.