World #1 Go Player Ke Jie accepts Google Alpha Go Match..

[duncan]

possibly AlphaGo will meet deep blue in chess heaven and have a fight over who is smarter.

pity. it could give handicaps to lee sedol and see if it can still win


https://techcrunch.com/2017/05/27/googl ... -retiring/

Google’s AlphaGo — the AI developed to tackle the world’s most demanding strategy game — is stepping down from competitive matches after defeating the world’s best talent. The latest to succumb is Go’s top-ranked player, Ke Jie, who lost 3-0 in a series hosted in China this week.

[Laskos]

I did not see a convincing evidence for go.

Note that I do not deny the fact that rating gap between beginner and best players are bigger in go relative to chess but it is evidence for nothing.

I can define a new game.
Chess*10

To play chess*10 you need to play a match of 10 chess games against another player.
draw is only if the result is 5-5 and if you score at least 5.5 you win.

The rating gap between beginner and world champion in chess*10 is clearly higher than the rating gap between beginner and world champion in chess but the best players in chess*10 are also the best players in chess.

Browsing through Go rating schemes, I remembered this post.

Chess*10 dilates ELO ratings by about a factor 2.4 across wide range of ELO span. Chess*10 seems a bit exaggerated compared to Go. The number of moves is only a factor of 3 larger in Go. If we take Chess*4 (more reasonable), then ELO in Chess will dilate by a factor of 1.4.

About Go ratings, here are useful links:
https://senseis.xmp.net/?GoR
https://www.goratings.org/en/
https://en.wikipedia.org/wiki/Go_ranks_and_ratings

Going to usual logistic ELO with a=400 (in Go they are using logistic, but in various and variable shapes), I calculated our ELO, for difference between 10 kyu intermediate amateur (this is at least 1200 FIDE ELO in Chess, as skill, talent, time consumed and preparation go) and a good Pro player, it came at 3400 usual ELO points. Ke Jie has another at least 400 (usual) ELO points, so at least 3800 normal ELO points span between Ke Jie and 10k. In Chess, we have 1600 ELO points between 1200 player and Carlsen. A factor of 2.4 larger ELO points span in Go.

Another take: the difference between top 1000 Chess players is about 350 ELO points. The difference between top 1000 Go players is 1100 ELO points. This is a factor of 3.1 larger in Go.

So, either top Go players are more skilled at Go than top Chess players at Chess, or one game of Go is indeed equivalent to 10-15 games of Chess in the sense of separating power between humans, which in itself would be remarkable.

[duncan]

any handicap matches against go world champions available ?
http://www.bbc.co.uk/news/technology-41668701


This has turned out to be far more efficient way of addressing the problem.

Whereas AlphaGo took months to get to the point where it could take on a professional, AlphaGo Zero got there in just three days, using a fraction of the processing power.

"It shows it's the novel algorithms that count, not the compute power or the data," says Mr Silver.

He enthuses about an idea some may find rather scary - that in just a few days a machine has surpassed the knowledge of this game acquired by humanity over thousands of years.

"We've actually removed the constraints of human knowledge and it's able, therefore, to create knowledge itself from first principles, from a blank slate," he said.


Whereas earlier versions quickly learned from and improved upon human strategies, AlphaGo Zero developed techniques which the professional player who advised DeepMind said he had never seen before.

Many of the team have now moved on to new projects where they are trying to take this technique to new areas. Demis Hassabis mentions drug design and the discovery of new materials as areas of interest.

Whereas some see a threat from AI, he looks ahead with optimism.

"I hope these kind of algorithms will be routinely working with us as scientific experts medical experts on advancing the frontiers of science and medicine - that's what I hope," he says.

But he and his colleagues are cautious about how rapidly we will see the wider application of these AI techniques - a game with clear rules and no element of luck is one thing, the messy, random, unpredictable real world quite another.

I wrote earlier this week about the tidal wave of AI hype pouring into my email inbox. AlphaGo Zero is at the other end of the spectrum - proper peer-reviewed science with a real advance in computer intelligence.