Go has fallen to computer domination?

[mar]

The following is simply nonsense.

You made my day Uri, thanks. I always enjoy your posts, more than Henk's - because you're actually serious

[towforce]

Most of the great men of computer chess are "enthusiastic amateurs" (EAs). It seems to me that AlphaGo marks the beginning of the end of the EA in AI: from now on, it will be all about finding the best way to:

a. make a trainable system that's suitable for learning a particular skill

b. run a training process

These are going to be big tasks that will require lots of resources - like the art of making a movie became in the early 20th century, and making a video game became in the late 20th century.

There are still things EAs can do in AI - but it seems unlikely they'll be doing many "breakthrough" tasks.

[Isaac]

Unfortunately, since the broadcast is Flash based I can't watch it.

I don't have Flash and I can still watch the broadcast live. In linux I open a terminal and type "mpv youtube-url" and I'm done.

[duncan]

http://www.theguardian.com/technology/2 ... go-contest

and what is a top computer game without it playing a move humans can never make

[Dirt]

Unfortunately, since the broadcast is Flash based I can't watch it.

I don't have Flash and I can still watch the broadcast live. In linux I open a terminal and type "mpv youtube-url" and I'm done.

I'll give that a try next game. Thanks!

[chetday]

AlphaGo wins first match. Very interesting article at

http://www.theverge.com/2016/3/9/111850 ... nce-impact

[Evert]

You know, it's actually not so interesting whether AlphaGo can beat the best human. What's interesting is what insights it can generate about the game itself.

There's a neural net, which encode a wealth of information in a way that is very hard to access and interpret. Tapping into that knowledge is far more interesting than the question of whether a black-box neural net can beat a human.

A similar point holds for Giraffe: what is interesting about it is not how strong it is, but figuring out what evaluation patterns it has learned and what this tells us about the game. I don't suppose anyone has even attempted that (but I haven't read the literature, so this is pure speculation on my part) and personally I'm not even sure how to go about something like that.

EDIT: at the same time, I don't think many of us here are under any sort of delusion that what we do has anything at all to do with AI or AI research.

[Milos]

EDIT: at the same time, I don't think many of us here are under any sort of delusion that what we do has anything at all to do with AI or AI research.

Today AI is all about the pattern recognition, something that birds even insects can do equally well.
What we do has nothing to do with bird or insects, so...

[chetday]

You know, it's actually not so interesting whether AlphaGo can beat the best human. What's interesting is what insights it can generate about the game itself.

I don't know much about Go, but in the article link I posted above, I found these two paragraphs especially interesting:

But AlphaGo could also open up new avenues for the game. Members of the Go community are as stunned with the inventive, aggressive way AlphaGo won as the fact that it did at all. "There were some moves at the beginning — what would you say about those three moves on the right on the fifth line?" American Go Association president asked VP of operations Andrew Jackson, who also happens to be a Google software engineer, at the venue following the match. "As it pushes from behind?" Jackson replied. "If I made those same moves…" Okun continued. "Our teachers would slap our wrists," Jackson agreed. "They’d smack me!" says Okun. "You don’t push from behind on the fifth line!"

"We’re absolutely in shock," said Jackson. "There’s a real question, though. We’ve got this established Go orthodoxy, so what’s this going to reveal to us next? Is it going to shake things up? Are we going to find these things that we thought were true — these things you think you know and they just ain’t so?"

[bob]

You know, it's actually not so interesting whether AlphaGo can beat the best human. What's interesting is what insights it can generate about the game itself.

I don't know much about Go, but in the article link I posted above, I found these two paragraphs especially interesting:

But AlphaGo could also open up new avenues for the game. Members of the Go community are as stunned with the inventive, aggressive way AlphaGo won as the fact that it did at all. "There were some moves at the beginning — what would you say about those three moves on the right on the fifth line?" American Go Association president asked VP of operations Andrew Jackson, who also happens to be a Google software engineer, at the venue following the match. "As it pushes from behind?" Jackson replied. "If I made those same moves…" Okun continued. "Our teachers would slap our wrists," Jackson agreed. "They’d smack me!" says Okun. "You don’t push from behind on the fifth line!"

"We’re absolutely in shock," said Jackson. "There’s a real question, though. We’ve got this established Go orthodoxy, so what’s this going to reveal to us next? Is it going to shake things up? Are we going to find these things that we thought were true — these things you think you know and they just ain’t so?"

Computers have shaken traditional thinking from day one. For example, "in a KQ vs KR endgame, the weaker side MUST keep the rook close to the king to prolong the loss as long as possible." Wrong.

Things are only obvious once they are shown to be true. Who knows what will become of this. And if this will apply to chess, things will get even more interesting...