AlphaGo retires from competition after humiliating humankind !
As expected, during the AI summit in Wuzhen (cf. https://www.gofed.be/node/1036), AlphaGo (AG) swept all its human opponents: Ke Jie, the best player in the word (3-0 !) and the team composed by 5 top Chinese professional (1-0).
Unexpected was the announcement by DeepMind that this games were the last ones AG had played and that they hence turn their interest to other domains than go...(the question "how many handicap stones a top pro would need to beat it" will thus remains open! :-(
DeepMind won't (unfortunately) communicate AG's program, but they will publish an article in "Nature" (a famous scientific review).
They also have released 50 games played by AG against itself...
http://www.go4go.net/go/alphago
Most of them are full of creative/ strange moves...In this video, Hwang In-seong (8d) showed the revolution in the way of thinking AG has brought (high level,in English!): https://www.youtube.com/watch?v=iO_JmGH8Iu8
More info here:
https://en.wikipedia.org/wiki/Future_of_Go_Summit
https://en.wikipedia.org/wiki/AlphaGo
http://361points.com/articles/thoughtsonalphago/
Many articles were published during this event:
among dozens of them:
https://www.theverge.com/2017/5/30/15712300/alphago-ai-humanity-google-a...
A extended press review (in French): http://rfg.jeudego.org/item/320-wuzhen-ou-le-dernier-tournoi-d-alphago