AlphaGo vs Tygem - Alpha as White - Master(P)

[@Mooskym]

6:15 "This stupid little computer is killing everybody"... That is the single scariest thing I have ever heard on the internet!

[@skidelrymar]

lol

[@AndrewInmanartist]

I watched one video of two alpha go games called "Alphago - Whatever you do is wrong" by Brady Daniels. My conclusion is Alphago will always have the advantage if humans are unable to settle a group with sente all the time. With the processing capabilities of Alphago, it can determine by positional judgement and counting what's really sente or not. As seen in both games in his video there was two groups that could be attacked...which happens between humans a lot because either they are unaware that the group isn't safe or they are trying to be aggressive to keep sente. People today prefer exciting fighting games. Alphago plays more traditionally. If you review older games, doesn't have to be Shusaku era without komi, Black always just had to maintain composure and keep the lead. White plays a bit aggressive to take the lead but then just maintains the lead, like in this game Alphago built influence in sente and maintained the lead.

[@TamakiFox]

I love watching these go videos even though I suck at go and have no idea what's going on nor the meaning of each move T.T

[@CAustin582]

Now that AlphaGo exists, we all suck at go, relatively speaking.

[@adambirch6466]

I love the historical legends games! The shape of their play is so gorgeous.

[@themanwiththepan]

I want to see alphago as white play takemiya masaki

[@BradyDaniels1]

Thanks for your AlphaGo videos. I, for one, am sad you're done. One element of your presentation style that I really like is when you play only one color's stones, inviting the viewer to imagine the other's to understand the sequence. You did that twice here with double hanes. FWIW, according to a list posted to Reddit, Black was Liu Yuhang, a brand new Chinese pro.

[@dwyrin]

Don't be sad! There are so many games played each week that a lot of really good material is missed if one only concentrates on AlphaGo. To give a few example of some candidates on my list; Mok Jinseok had a game of jumping battles that cut to the heart of why that extra jump is important, Kono Rin played against a super crazy huge framework for a winner-takes-all reduction, Shibano Toramaru 3p had to play against an 8p in the Taiwan-Japan exchange matches for a risky game of influence, etc.. The list just continues to grow and I really want to cover some of these really fun games even if they aren't as click baity as the AlphaGo hype.

[@BradyDaniels1]

dwyrin I look forward to watching those games too, even if I believe that AlphaGo is a much bigger deal than clickbait. Have fun, and keep doing what you're doing.

[@dwyrin]

Eh? I didn't say that at all. I said the only reason I would focus on alphago videos to the exclusion of all other games and all other players, especially given how many people are already doing so, would be for the views the name generates rather than the educational and entertainment value and diversity said other games would bring.

[@BradyDaniels1]

dwyrin Cool. I agree and look forward to whatever you do.

[@Effivera]

Dear Brady, I love Dwyrin's alphago videos and wished he would do more as well. Having said that, I also loved your own alphago videos - I wish you would tackle a few of the Master games for our viewing pleasure. Thank you both for great channels.

[@jejeakle]

And once again the corner proverb proves right

[@thembamabona9809]

....as a beginner i have to say, though I do wish to make progress, I am much more reluctant now to learn Josekis before finding out which ones AlphaGo has proven to be more or less useless versus those that continue to make good sense... is this a common occurrence?

[@szabolcsgal4188]

Don't worry about verified and non verified Josekis, I would say don't really worry about Josekis at first. Just play the game however you enjoy it, and later you would have an urge to learn Josekis anyway. (And at that point they are much easier to learn, because the moves will actually start to make sense.) Also everyone should be a bit doubtful on the AlphaGo verified Josekis/moves. They might be worthwhile for professionals, but there could be moves that are just out of human reach. For example, there might be a very sharp move, that works if you can read every variation 30 moves ahead, but if you are unable to do so, it might loose you the game. Obviously it's not worth making that move if you can't read 30 moves ahead. Now the problem comes if that move is really sharp, but the reasons for that only surfaces after 20 moves. Human mind simply might not be able to see that far ahead, and will end up following something the computer does blindly, where a simple, but safer move would have been better. This applies for almost any games for almost with large skill difference btw. There might be strategies, which are superior to some strategy, but is so hard to pull off / too glass cannon like, that it's not worth to go for below a certain skill level. Applying this to Josekis: there is a lot of information on how hard to play a Joseki is, or Joseki for beginners, not just objectively how effective it is.

[@bernardfinucane2061]

Black got all four corners. Good enough for second place I guess.

[@mecemodzada8315]

Black is supposed to be Liu Yuhang and he is not a pro player.

[@dwyrin]

As pointed out below he is in fact a new professional 1p You can see his games here - www.go4go.net/go/games/byplayer/1582

[@mecemodzada8315]

Nice, thanks.

[@TheFlowrider7]

so that stupid bot is not loosing a single match? :(