I find this very disappointing...
![]()
[Wired]Humans No Match for Go Bot Overlords
1 - 20 of 23 Posts
2
Quote:
You thought this was going to be about the 'Wal-mart Transformers' too, huh?
Very disappointed...
Originally Posted by Solertia I find this very disappointing...
|
Very disappointed...
Quote:
They're not doing a "brute force" type approach. Go is too complicated for a "if-then" sequence to achieve any moves of value.
Rather, as described in the article, they are letting the computer run millions of random games - just filling in the board with no purpose or plan - and then using statistics and probability to 'teach' the computer how to play. It's a pattern-finding method, more heuristic and subtle than simply building a search tree.
They are not surprised that the computer can do the calculation, but that the computer can learn anything from playing randomly. And not just learn, but learn effectively.
Something similar was discovered a few years back, albeit somewhat inverted. Researchers gave a simple rule-set to a computer to produce patterns, and they found a rule-set that yielded what appeared to be completely random outputs. It's fascinating stuff, not well understood.
Originally Posted by Coma But this doesn't do it like a human, it does it better. I don't see why they're surprised it works - no AI is needed to play something like Go, just pure logic. |
Rather, as described in the article, they are letting the computer run millions of random games - just filling in the board with no purpose or plan - and then using statistics and probability to 'teach' the computer how to play. It's a pattern-finding method, more heuristic and subtle than simply building a search tree.
They are not surprised that the computer can do the calculation, but that the computer can learn anything from playing randomly. And not just learn, but learn effectively.
Something similar was discovered a few years back, albeit somewhat inverted. Researchers gave a simple rule-set to a computer to produce patterns, and they found a rule-set that yielded what appeared to be completely random outputs. It's fascinating stuff, not well understood.
Quote:
Cars already drive themselves...
Originally Posted by TestECull Cool. When they can concoct a robot that can drive any automobile just as a human does, I'll be impressed...and downright scared too, considering the dumbass moves I've seen real humans do. There's no telling what kind of idiotic stunts a robot might pull, seeing as robots don't have fear... |
Quote:
Yes, but can they run red lights and get into an accident because they were drunk and wanted to change the cd? I don't think so.
Originally Posted by Licht Cars already drive themselves... |
Quote:
No, they don't...not on public highways. Right now, robot cars are still too stupid to not hit anything unless they implant a track in the roadway for them to follow, and then they're no more complex than a 10 dollar line tracking mouse kit for electronics 1 students in high school.
The only self driving cars that don't hit things and don't follow a track are merely 1:1 scale RC's piloted by humans standing on rooftops nearby...
Originally Posted by Licht Cars already drive themselves... |
No, they don't...not on public highways. Right now, robot cars are still too stupid to not hit anything unless they implant a track in the roadway for them to follow, and then they're no more complex than a 10 dollar line tracking mouse kit for electronics 1 students in high school.
The only self driving cars that don't hit things and don't follow a track are merely 1:1 scale RC's piloted by humans standing on rooftops nearby...
Quote:
No they use cameras to follow the lane markings that already exist. Why would they need something in the middle of the road when we humans have something there already for ourselves?
Originally Posted by TestECull No, they don't...not on public highways. Right now, robot cars are still too stupid to not hit anything unless they implant a track in the roadway for them to follow, and then they're no more complex than a 10 dollar line tracking mouse kit for electronics 1 students in high school. The only self driving cars that don't hit things and don't follow a track are merely 1:1 scale RC's piloted by humans standing on rooftops nearby... |
3
Quote:
Quote:
If it makes you guys feel any better, computers process data way faster than we can, and the only way this can be made to work is because computers are so good at brute forcing their way through problems.
Crunching through millions of games of Go just to finish 1 game? If you lose I guess you can just say the computer's been playing way more than you, heh.
Originally Posted by Solertia I find this very disappointing...
|
Quote:
Originally Posted by Ruei ... Wow... I think a part of me just died a little. |
Crunching through millions of games of Go just to finish 1 game? If you lose I guess you can just say the computer's been playing way more than you, heh.
Quote:
This is somewhat inaccurate. Human brains (and down the food chain to at least insect nervous systems) process massive amounts of information fairly quickly. Visual perception for example; the mind can take in constant, continually varying and very complicated environmental input and make almost instant sense of it. Or, if you prefer in tech-geek-speak, render a massive resolution (better than any monitor) 3D image in real-time, and even perform complicated image enhancement algorithms on the fly.
Add sound, touch, smell, taste, temperature regulation, breathing, heart beating, and being able to engage in conscious thought on top of it all, is very impressive computational architecture.
It's important to keep in mind, computers do one thing exceptionally well - exactly what we tell them to do (to the chagrin of many an unclear programmer). That they may do that fast isn't all too impressive on its own. The equation itself is just consistent gibberish, the thought behind it is what is inspiring.
Originally Posted by Arkanor If it makes you guys feel any better, computers process data way faster than we can, and the only way this can be made to work is because computers are so good at brute forcing their way through problems. Crunching through millions of games of Go just to finish 1 game? If you lose I guess you can just say the computer's been playing way more than you, heh. |
Add sound, touch, smell, taste, temperature regulation, breathing, heart beating, and being able to engage in conscious thought on top of it all, is very impressive computational architecture.
It's important to keep in mind, computers do one thing exceptionally well - exactly what we tell them to do (to the chagrin of many an unclear programmer). That they may do that fast isn't all too impressive on its own. The equation itself is just consistent gibberish, the thought behind it is what is inspiring.
Quote:
While that is true, the human brain is limited in that respect. It's very good at what it does, but computers can crunch any data set we tell them to. And they're only going to get faster at doing it. That's the reason brute force approaches are workable on computers where they would be completely ineffective using human brains.
The power of the human brain in solving problems comes in fuzzy logic, pattern recognition, and intuitive approaches (elegant methods to solving problems that avoid resorting to brute force.) If for example you create a display containing all red squares and 1 blue square, a person could immediately tell you where the blue square was. If you asked them how they did it they would probably just look at you funny and say they just saw it.
To do the same thing on a computer, you would need it to test every square every time, and isolate the one that was different. However, since it can do all of this so fast, this 'just-crunch-it-all' approach on a computer would probably return your result faster than a person.
Originally Posted by dharmaBum This is somewhat inaccurate. Human brains (and down the food chain to at least insect nervous systems) process massive amounts of information fairly quickly. Visual perception for example; the mind can take in constant, continually varying and very complicated environmental input and make almost instant sense of it. Or, if you prefer in tech-geek-speak, render a massive resolution (better than any monitor) 3D image in real-time, and even perform complicated image enhancement algorithms on the fly. |
The power of the human brain in solving problems comes in fuzzy logic, pattern recognition, and intuitive approaches (elegant methods to solving problems that avoid resorting to brute force.) If for example you create a display containing all red squares and 1 blue square, a person could immediately tell you where the blue square was. If you asked them how they did it they would probably just look at you funny and say they just saw it.
To do the same thing on a computer, you would need it to test every square every time, and isolate the one that was different. However, since it can do all of this so fast, this 'just-crunch-it-all' approach on a computer would probably return your result faster than a person.
1 - 20 of 23 Posts
