Google's AI Masters Go Board Game A Decade Early

Google's AI Masters Go Board Game A Decade Early

Google's artificial intelligence can beat expert Go players ten years earlier than science predicted.

In just about any science fiction movie, a scientist saying "They're learning faster than expected" is cause for alarm. And while we shouldn't break out EMPs and ion weapons, that's exactly what the artificial intelligence community is saying about Google's AI. It seems this computer system doesn't just play Go, it's also good enough to surpass expert human players. Considering most programmers believed this wouldn't be possible for another ten years, that's kind of a big deal.

The AI in question is AlphaGo, a computer program that just beat European Go champion Fan Hui at his own game. "Go is the most complex and beautiful game ever devised by humans," Google's Demis Hassabis said, adding that AlphaGo's victory "achieved one of the long-standing grand challenges of AI."

Okay, so AlphaGo won a board game. What's the big deal? Don't computers win at chess all the time?

The truth is Go is a far more complex game than chess ever could be. Go games typically last 150 moves, with an average 250 choices per move. Complicating matters is the fact that victory depend on recognizing subtle patterns in the arranged pieces, which is far harder for computers than it sounds. Even human Go champions can't always verbalize how to take advantage of certain patterns, so AI experts thought we needed at least a decade before computers could do the same.

What makes AlphaGo so special compared to other AI? It's apparently the use of two deep-learning networks - one to predict moves and another to predict outcomes. These are then combined under traditional AI algorithms to produce results which a computer could quickly grasp. "The game of Go has an enormous search space, which is intractable to brute-force search," Google researcher David Silver explains. "The key to AlphaGo is to reduce that search space to something more manageable. This approach makes AlphaGo much more humanlike than previous approaches."

AlphaGo will face Lee Sedol, one of the best Go players in the entire world, in March to see how these algorithims hold up. Either way, it's an impressive achievement - one Google wishes to apply to humanitarian goals. "Ultimately we want to apply these techniques to important real-world problems," Hassabis said. "Because the methods we used were general purpose, our hope is that one day they could be extended to help address some of society's most pressing problems, from medical diagnostics to climate modeling."

I have to admit, solving climate change would be a pretty gracious way for the AI that crushed all our Go champions to behave. Here's hoping learning faster than expected proves to be a good thing.

Source: Nature, via Technology Review

Permalink

Fanghawk:
Google's AI Masters Go Board Game A Decade Early

Fanghawk:
Google's AI Masters

I knew it!

Not selling this accomplishment in robotics short, but AI still has a waaaay to go to master this game. This isn't like deep blue beating Kasparov, who was one of the best chess players to ever live.

The player in question, Fan Hui, is a professional 2 Dan go player. This makes him very strong, yes, but there are SEVEN levels of professional above him, eight if you count the title holders in Japan, China, and Korea, where the level of play is the highest.

Still, this is quite an impressive feat in programming, since winning an even game against a professional player at any level was science fiction for a long time

Edit: having said all that, their match with lee sedol is one I won't be missing

The real scary thing is how fast this kind of tech will be able to replace other service jobs pretty soon. Robotics for manual labor, AI for service jobs, this is getting real.

It means we will likely live during the transition between a human work force and an automatic work force. That transition can go any number of ways and a lot of them have significant human suffering until legislation kicks in to start forcing companies to either hire a certain number of humans or figure out some sort of way to keep the consumer workforce up (including just a tax based payout each year for the unemployed directly derived from proportionately increased taxes on automated businesses.

The eventual conclusion should me the death of "having to" work. But it's an uncertain walk in the meantime.

Lightknight:
The real scary thing is how fast this kind of tech will be able to replace other service jobs pretty soon. Robotics for manual labor, AI for service jobs, this is getting real.

It means we will likely live during the transition between a human work force and an automatic work force. That transition can go any number of ways and a lot of them have significant human suffering until legislation kicks in to start forcing companies to either hire a certain number of humans or figure out some sort of way to keep the consumer workforce up (including just a tax based payout each year for the unemployed directly derived from proportionately increased taxes on automated businesses.

What's probably going to happen is that other job fields are going to get more competitive as robots/computers push humans out of low level job types. I wonder if we will see an increase in schools focusing on skill sets distinct to humans like art or design?

Lightknight:
The real scary thing is how fast this kind of tech will be able to replace other service jobs pretty soon. Robotics for manual labor, AI for service jobs, this is getting real.

Not as much at all as you might think. Sure, some people will lose their jobs but on the other hand, having robots in place to do the stupid dead-end jobs will also free up more positions in higher places.

Arnoxthe1:

Lightknight:
The real scary thing is how fast this kind of tech will be able to replace other service jobs pretty soon. Robotics for manual labor, AI for service jobs, this is getting real.

some people will lose their jobs

In the US it's estimated at about 6 million jobs in the transportation industry alone.
It's quite a bit more then some. It's going to be a rater pressing issue within 10 years or so.

direkiller:
In the US it's estimated at about 6 million jobs in the transportation industry alone.
It's quite a bit more then some. It's going to be a rater pressing issue within 10 years or so.

And perhaps 6 million more jobs were created. I'm just saiyan. Let's not be all doom and gloom about this.

Read title as "Google's AI masters go overboard" for some reason. I joined suspecting to hear about some sort of proto-skynet. Was a little disappointed, but at least now I will know that it will likely not be bested ala Wargames style duel of game. We shall see what google's AI masters do next I suppose.

While being slowly consumed by robots sounds scary, if we consider their assimilation as the ultimate record of human existence, our deaths at the hands of the archivists is probably the longest lasting and best use of our legacy.

weirdee:
While being slowly consumed by robots sounds scary, if we consider their assimilation as the ultimate record of human existence, our deaths at the hands of the archivists is probably the longest lasting and best use of our legacy.

Are you the PR agent for Reapers? :p

LetalisK:

weirdee:
While being slowly consumed by robots sounds scary, if we consider their assimilation as the ultimate record of human existence, our deaths at the hands of the archivists is probably the longest lasting and best use of our legacy.

Are you the PR agent for Reapers? :p

i mean, what's the harder sell, a hundred babies, or a giant metal skeleton that can fire laser beams

 

Reply to Thread

Log in or Register to Comment
Have an account? Login below:
With Facebook:Login With Facebook
or
Username:  
Password:  
  
Not registered? To sign up for an account with The Escapist:
Register With Facebook
Register With Facebook
or
Register for a free account here