Computer Go at master level?

Discussion of chess software programming and technical issues.

Moderators: hgm, Rebel, chrisw

nionita
Posts: 175
Joined: Fri Oct 22, 2010 9:47 pm
Location: Austria

Computer Go at master level?

Post by nionita »

Anyone already read this article? Without subscription only the abstract is available...

http://www.nature.com/nature/journal/v5 ... 16961.html
User avatar
Ajedrecista
Posts: 1966
Joined: Wed Jul 13, 2011 9:04 pm
Location: Madrid, Spain.

Re: Computer Go at master level?

Post by Ajedrecista »

Hello Nicu:
nionita wrote:Anyone already read this article? Without subscription only the abstract is available...

http://www.nature.com/nature/journal/v5 ... 16961.html
I suggest search the title of the paper on Internet and see what happens. The abstract tells very good match results by the way.

This topic is also discuted here:

Go has fallen to computer domination?

Although I do not include pirate links, moderators can feel free of delete or edit my post if they think it is inappropriate.

Regards from Spain.

Ajedrecista.
User avatar
Evert
Posts: 2929
Joined: Sat Jan 22, 2011 12:42 am
Location: NL

Re: Computer Go at master level?

Post by Evert »

It may come out on arxiv.org in a couple of days, if the authors submit it there (which is the norm in some fields, but I don't know about computer sience speciically).
Because it's a nature paper it can't have been put up there before it came out.
User avatar
Bloodbane
Posts: 154
Joined: Thu Oct 03, 2013 4:17 pm

Re: Computer Go at master level?

Post by Bloodbane »

Functional programming combines the flexibility and power of abstract mathematics with the intuitive clarity of abstract mathematics.
https://github.com/mAarnos
nionita
Posts: 175
Joined: Fri Oct 22, 2010 9:47 pm
Location: Austria

Re: Computer Go at master level?

Post by nionita »

Indeed! Thanks,
Nicu
matthewlai
Posts: 793
Joined: Sun Aug 03, 2014 4:48 am
Location: London, UK

Re: Computer Go at master level?

Post by matthewlai »

Ajedrecista wrote:Hello Nicu:
nionita wrote:Anyone already read this article? Without subscription only the abstract is available...

http://www.nature.com/nature/journal/v5 ... 16961.html
I suggest search the title of the paper on Internet and see what happens. The abstract tells very good match results by the way.

This topic is also discuted here:

Go has fallen to computer domination?

Although I do not include pirate links, moderators can feel free of delete or edit my post if they think it is inappropriate.

Regards from Spain.

Ajedrecista.
It is now available on the DeepMind page. No pirating required!

http://deepmind.com/alpha-go.html

"Read about Alpha-Go here"
Disclosure: I work for DeepMind on the AlphaZero project, but everything I say here is personal opinion and does not reflect the views of DeepMind / Alphabet.
mcostalba
Posts: 2684
Joined: Sat Jun 14, 2008 9:17 pm

Re: Computer Go at master level?

Post by mcostalba »

matthewlai wrote: It is now available on the DeepMind page. No pirating required!
Firstly I would like to congratulate with you and your employer for this fantastic achievement. It is really an important day for neural net technology.

It would be interesting to see how NN fares against chess champions.

Of course I don't mean humans, but traditional chess engines.

Indeed I think that chess is basically a computer oriented game, not a human oriented game.

In the 60' it was picked up as an AI test, perhaps because was seen as a low hanging fruit that 'resembled' what humans were supposed to excel. Of course today this seems a very naive approach considering that humans compared to computers excel in totally different topics than playing chess, but at the time other "human kind" topics as natural language, vision, etc.. were probably too far away from the possibilities of the time and so the chess 'trick' seemed a nice marketing vehicle to attract some attention (and funds) in this new AI research thing.

Today we know that traditional chess engines, i.e. engines programmed with a well known bag of typical computer science algorithms can easily outperform the world champion, and even on a mobile (as long as the batteries last :-)

What we don't know is if an engine that leverages NN for evaluation and search heuristics can outperform the careful hand-engineered and well tuned features and terms of a standard chess engine.

I hope that a NN chess engine will beat Stockfish one day, this would be the little revenge that humans take against computers!
nionita
Posts: 175
Joined: Fri Oct 22, 2010 9:47 pm
Location: Austria

Re: Computer Go at master level?

Post by nionita »

But then, of course, the next level would be to train a NN to programm a chess engine written in "classical" way which again is much better than what a NN ever could reach ;-)
Henk
Posts: 7216
Joined: Mon May 27, 2013 10:31 am

Re: Computer Go at master level?

Post by Henk »

Maybe GO relies more on pattern matching or classification than chess where calculation is important. So GO is probably more suitable for Neural networks.
whereagles
Posts: 565
Joined: Thu Nov 13, 2014 12:03 pm

Re: Computer Go at master level?

Post by whereagles »

mcostalba wrote:Indeed I think that chess is basically a computer oriented game, not a human oriented game.
The combinatorial complexity of chess is indeed overwhelming. Even the best champions will always overlook something.

In smaller games like Bridge (52 cards) complexity can be kept under control. Plus it has a lot of judgement involved, which makes it much more human-oriented. I keep nagging my NN expert friend for us to take a shot at cracking it, but he says he's too busy :)