World's top Go player loses second match of five against Google AI

[TheBigOhMan]

do you understand the difference between human learning and machine learning?

I was thinking about this. I've personally have played many strategy games, particularly trading card games (I've played so far Yugioh!, Magic, Pokemon, and a bunch of lesser known games ) and I was thinking how I learned to play them.

The first thing of course it to keep the rules of the game in mind, how to play it, how to win and the such. I think that's the easiest part. The hard part is to learn the so-called "unspoken rules", those are more like guides instead of rules but if you follow them you can increase your chances of winning. Unspoken rules usually involve particular card combinations, or certain general strategies. And those are the ones you learn by seeing how others play and by practice.

As you start practicing and seeing you start learning the unspoken rules. So if I play a lot, and I get my ass handed to me because the dude uses a particular card strategy, I add to my mind the rule "when playing use this and that card strategy". Of course, then I use it and someone developed a counter-strategy to that card stategy and so the rule becomes more complex like "when playing use this and that card strategy except when my oponent has this counter-strategy in which case use this and that card strategy". And so on and so forth.

I've also played against computers in the past, old programs from the 90s or mid 2000. They are quite easy to beat because they never learn, you can beat them over and over again using the same dumb strategies and the program will never catch the unspoken rule. it sometimes gets annoying because it's like playing with a stubborn child that just refuses to accept that he should change the focus of it's strategy.

Reading the thread, it seems the machine is an advancement from this old programs. The programmers only gave the written rules of Go, and the machine learned through observation the unwritten ones, so it's constantly changing. You may beat it the first time in a way, but it will probably not fall for the same trick twice.

[aleph naught]

I've worked for decades with various network typologies and search methodologies including bayesian networks and DAGs, they are all variations on the basic notions.

To the layman the statement "Games are won by massive search trees. Computers will always be able to do this better. No news. No surprise."  is absolutely correct. It is also specifically correct in the case of AlphaGo due to the use of Monte Carlo.

I agree with kurro, that's absolutely misleading to the lay-man. Even humans perform search trees when strategizing.

do you personally understand the difference between a search tree and a neural network? That would mean, can you articulate the fundamental difference without googling?

The fundamental difference?! They're nothing like each other. It's like you're asking me for the fundamental difference between a car an an apple. One is just a smart way of ordering objects so you can find what you're looking for in O(n log n) time. The other is a universal function approximator.

obfuscation! man, you sure love em..

both have decision networks and both have transform functions (see red–black tree for example)

You keep using that word, but I don't think you know what it means. My post was the exact opposite of obfuscation. I'm not even using complicated jargon, you should know exactly what I'm talking about if you have even a slight understanding of the topic.

Neverthesless, cars and apples are both made out of particles. That doesn't mean they're anything alike. It's becoming more and more apparent that you've got no clue what you're talking about.

But I'll ask you: what do you think the 'fundamental difference' is between a search tree and a neural network?

You haven't responded to what I said: even humans perform tree searches when strategizing. That the AlphaGo program uses them doesn't mean that's all it is. If it was, it wouldn't be able to win games against masters like it is doing. It is only able to win because it has a neural network that has learned how to play the game at the level of a master. In other words, a massively impressive feat of machine learning. It's learned things that most humans couldn't learn, or would take a very long time to learn if they could.

do you understand the difference between human learning and machine learning?

I don't know how human brains learn, and I expect no one really does. Do you?

And there isn't any single form of machine learning, so your question just really doesn't make sense.

[RichardChad]

I've worked for decades with various network typologies and search methodologies including bayesian networks and DAGs, they are all variations on the basic notions.

To the layman the statement "Games are won by massive search trees. Computers will always be able to do this better. No news. No surprise."  is absolutely correct. It is also specifically correct in the case of AlphaGo due to the use of Monte Carlo.

I agree with kurro, that's absolutely misleading to the lay-man. Even humans perform search trees when strategizing.

do you personally understand the difference between a search tree and a neural network? That would mean, can you articulate the fundamental difference without googling?

The fundamental difference?! They're nothing like each other. It's like you're asking me for the fundamental difference between a car an an apple. One is just a smart way of ordering objects so you can find what you're looking for in O(n log n) time. The other is a universal function approximator.

obfuscation! man, you sure love em..

both have decision networks and both have transform functions (see red–black tree for example)

You keep using that word, but I don't think you know what it means. My post was the exact opposite of obfuscation. I'm not even using complicated jargon, you should know exactly what I'm talking about if you have even a slight understanding of the topic.

Neverthesless, cars and apples are both made out of particles. That doesn't mean they're anything alike. It's becoming more and more apparent that you've got no clue what you're talking about.

typical from you, try the semantic route, try the obfuscation route, personal insult, exit thread.

One of the two of us works in the field of computational analytics professionally.

But I'll ask you: what do you think the 'fundamental difference' is between a search tree and a neural network?

For the layman who doesn't even know what a BST is? Who is just using the word "tree" to indicate an ability to analyze different routes? Not much at all, and that's the entire point, its just a smarter way to analyze more options.

You haven't responded to what I said: even humans perform tree searches when strategizing. That the AlphaGo program uses them doesn't mean that's all it is. If it was, it wouldn't be able to win games against masters like it is doing. It is only able to win because it has a neural network that has learned how to play the game at the level of a master. In other words, a massively impressive feat of machine learning. It's learned things that most humans couldn't learn, or would take a very long time to learn if they could.

do you understand the difference between human learning and machine learning?

I don't know how human brains learn, and I expect no one really does. Do you?

And there isn't any single form of machine learning, so your question just really doesn't make sense.

lol, semantics again.
Computers learn by having the ability to incorporate more and more information into a deterministic data analysis algorithm. The speeds increase, and the information pool increases, and the deterministic data analysis algorithm gets better and better, but that's light years from having the ability to reason, deliberate and choose otherwise.

[Lion IRC]

Right.
This cold, robotic indifference makes the program conspicuously LESS human.

Of course AlphaGo is conspicuously nonhuman.  If you think anyone in this thread has suggested otherwise then that is just something else you have misunderstood.

Great! Everyone agrees. Glad we had this talk.

...It is kind of amazing to me that you guys are so threatened by this, that you need to engage in so much denial.

You clearly missed my point about the Ferrari versus Usain Bolt.

I don't think Usain Bolt feels 'threatened' by a car.

Neither am I surprised a computer beat a human at a chess game or a Go game.

What's amazing is that a human can beat the computer.
/thread