AlphaGo beats Lee Sedol again in match 2 of 5 • Hacker News Board

Re: AlphaGo beats Lee Sedol again in match 2 of 5

#101

Post by tzs » Thu, Mar 10, 2016, 10:08 AM UTC

sidko wrote: post #73

imron wrote: post #62

Earlier quoted context omitted.

I wonder how long before humans start learning from AlphaGo! I want to see AlphaGo vs AlphaGo :-)

Seeing the way chess computers have evolved, this won't be far into the future. There are several tournaments just for chess computers, and it is very hard even for grandmasters to follow every more. Somehow, after a 20-move sequence that appears to accomplish nothing, one side is slightly up, and the rest of the game is decided.

> Seeing the way chess computers have evolved, this won't be far into the future.

With chess, there were two breakthroughs. First, there was Deep Blue, which threw massive hardware resources at the problem and achieved world champion level play.

That was interesting, of course, but didn't really do anything for human chess, because most humans did not have access to the necessary hardware.

The second breakthrough was when the developers of chess programs that ran on commodity desktop computers improved their algorithms to the point that they could play at (and far beyond) Deep Blue's level even though they were only able to search about 1/100th as many positions per second.

That was when humans started being able to really use computers to help the humans understand chess.

The breakthroughs in chess algorithms on commodity computers had little, if anything, to do with the Deep Blue breakthrough. The two are just too different.

Can AlphaGo be made available on hardware that top human Go players have access to, or is AlphaGo to Go as Deep Blue was to chess?

Re: AlphaGo beats Lee Sedol again in match 2 of 5

#102

Post by trampi » Thu, Mar 10, 2016, 10:08 AM UTC

albertzeyer wrote: post #59

pushrax wrote: post #3

If AlphaGo wins all 5 matches, what do you think DeepMind will do with it? My intuition is that they won't continue development, and instead focus on other applications. Great game btw, a pleasure to watch.

StarCraft is their next challenge. :)

Re: AlphaGo beats Lee Sedol again in match 2 of 5

#103

Post by School-Cotton » Thu, Mar 10, 2016, 10:08 AM UTC

Scea91 wrote: post #90

KennyCason wrote: post #66

Earlier quoted context omitted.

I encountered this many times over the past couple of days haha. I then have to explain that previous AIs, like those in chess largely used brute force computation to simulate each move, while these AIs "actually learn, similar to our brain". Probably not the most scientific explanation, but I feel your sentiment.

There is still tree search going on. However I don't know the details about it. Is it that the deep net works mainly as an evaluation function of the current position? I guess it does more than that right?

There's also tree search in my brain when I play chess or go. (Consciously!)

Re: AlphaGo beats Lee Sedol again in match 2 of 5

#104

Post by Jach » Thu, Mar 10, 2016, 10:10 AM UTC

kqr wrote: post #83

Jach wrote: post #78

Earlier quoted context omitted.

I had this conversation earlier with a friend wondering if any of Sedol's Korean pro buddies have noticed any systematic biases that could be exploited. I think it would be possible to make the neural net relatively useless by playing strange sequences that hack the weights, but you're still left with a monte-carlo-tree-search bot which alone depending on its implementation is between 2d and 6d amateur (so on the far…

I'm fairly certain that among the many thousands of high-level games from KGS AlphaGo was trained on, several opened tengen and 5,5.

Sure, almost any HnG fan will play them at least once. ;) But there's so few of them in general I would think any training data derived from them would be low-value.

Re: AlphaGo beats Lee Sedol again in match 2 of 5

#105

Post by mmahemoff » Thu, Mar 10, 2016, 10:11 AM UTC

skc wrote: post #49

I find it very interesting that to a layperson, the idea of a computer being able to beat a human at a logic game is pretty much expected and uninteresting. You try and share this story with a non-technical person and they will likely say "Well, duh..it's a computer".

Most people operate on a moving definition of intelligence as "whatever humans can do and machines can't" (thus ruling out AI by definition).

If software started writing bestselling novels, it would soon become a "duh that's just what computers do" matter of fact.

Re: AlphaGo beats Lee Sedol again in match 2 of 5

#106

Post by spdy » Thu, Mar 10, 2016, 10:13 AM UTC

Must be amazing seeing how the program you helped to create beat the best player in one of the most complex games on this planet.

This is a milestone in modern informatics.

Re: AlphaGo beats Lee Sedol again in match 2 of 5

#107

Post by kqr » Thu, Mar 10, 2016, 10:13 AM UTC

derefr wrote: post #82

kqr wrote: post #77

Earlier quoted context omitted.

I can't find the source now, but I remember reading a while ago that a team of go players is, perhaps contrary to intuition, not significantly better than their strongest individual alone. There's just no useful way to "pool" human thinking power and redistribute it to where it's needed the most – all the players will simultaneously consider the same branches. At best you reduce the risk of making a silly mistake, bu…

> all the players will simultaneously consider the same branches A computer-assisted pool of humans might work. Feed each human a board state advanced by 1/2/3/N moves down the decision tree in some direction, and have them evaluate that particular sub-tree. It's a map-reduce problem!

That's actually a really interesting idea. At this point you're basically replacing the "policy" neural network in AlphaGo with biological human neural networks.

Re: AlphaGo beats Lee Sedol again in match 2 of 5

#108

Post by i_don_t_know » Thu, Mar 10, 2016, 10:16 AM UTC

sago wrote: post #58

i_don_t_know wrote: post #42

Is there a complete recording of the commentary? They had one for game 1. The current live stream only goes back two hours and doesn't include the beginning of the game. I'm looking at the DeepMind channel on Youtube: https://www.youtube.com/channel/UCP7jMXSY2xbc3KCAE0MHQ-A

They put a version of the broadcast up after it is finished, if last time is anything to go by. Took a little while because the stream is still going, with the press conference.

Thank you. The full six hour version is available now.

Re: AlphaGo beats Lee Sedol again in match 2 of 5

#109

Post by vedaprodarte » Thu, Mar 10, 2016, 10:23 AM UTC

A question: Should we take it as "a computer beating a human" or "developers beating a Go player"? I had this discussion with my friends and we have opposite opinions.

Re: AlphaGo beats Lee Sedol again in match 2 of 5

#110

Post by jacquesm » Thu, Mar 10, 2016, 10:24 AM UTC

tzs wrote: post #101

sidko wrote: post #73

Earlier quoted context omitted.

Seeing the way chess computers have evolved, this won't be far into the future. There are several tournaments just for chess computers, and it is very hard even for grandmasters to follow every more. Somehow, after a 20-move sequence that appears to accomplish nothing, one side is slightly up, and the rest of the game is decided.

> Seeing the way chess computers have evolved, this won't be far into the future. With chess, there were two breakthroughs. First, there was Deep Blue, which threw massive hardware resources at the problem and achieved world champion level play. That was interesting, of course, but didn't really do anything for human chess, because most humans did not have access to the necessary hardware. The second breakthrough was…

The fact that Lee Sedols hardware is a couple of pounds of wetware running on a peanut butter sandwich suggests the answer to your question is yes.

Whether those insights will come soon or not is the big question.