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

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

#391

Post by criddell » Thu, Mar 10, 2016, 4:34 PM UTC

awl130 wrote: post #351

criddell wrote: post #332

Earlier quoted context omitted.

I'm not sure what will happen. Creative works aren't generally fungible. If I order tickets to go see Jack White play, I'm not going to be tempted by Nickelback tickets that are half the price.

you're confusing performance with the underlying intellectual property (e.g., the published song)

It's the same fungibility argument though. The market for sheet music for Willie Nelson's catalog isn't affected by the availability of other music, is it? If I want to play On the Road Again, there's only one IP owner for that.

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

#392

Post by luso_brazilian » Thu, Mar 10, 2016, 4:35 PM UTC

lmm wrote: post #244

crispyambulance wrote: post #215

Earlier quoted context omitted.

I suppose in the era of Bobby Fischer, chess was proxy for superpower one-upmanship. That's long gone, but Chess as a game is still doing fine and I expect that it will be the same for Go. We still have chess tournaments, super-star grandmasters and circus freaks (people who can play blindfolded against multiple opponents). And, yes, computers can easily smoke all but elite players. Why should Go be different?

I played Chess as a kid. I watched the local tournament shrink from the big town hall, to the side room in the same building, to a local school hall, to a classroom in that school. I really do think the game is dying - perhaps something that was happening already, but Kasparov losing to Deep Blue seemed to really catalyse it. (Not saying this is a bad thing. Evolution in games is natural, and I think it's amazing how…

I don't believe there is a relation between the improvement of computer chess and the reduction in the availability of in-person chess.

I also played chess as a kid and the allure of both local and national tournaments was that you could play with a multitude of different players, as opposed to the same 4 or 5 habitual chess players in your family/school/circle of friends.

But now, with the internet, at any second you can play with different people from all over the world, different strengths, styles and whatnot.

Hence now, instead of looking for the local chess club in the weekends we can play, any time of the day, any day of the week, anywhere.

Sites like the excellent lichess [1] are even free (in this case, free both as in beer but also as in speech) and, at any moment there there are 9 thousand, 10 thousand players enjoying this magnificent game.

[1] http://en.lichess.org/

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

#393

Post by soundwave106 » Thu, Mar 10, 2016, 4:35 PM UTC

imaginenore wrote: post #198

xiaq wrote: post #190

Earlier quoted context omitted.

Art is, mostly, domain of knowledge not yet claimed by science. When a technique turns art into science, it means that humans are ready to tackle even harder problems. Think about medicine before and after microscopes. Holy art turned into boring science: huge win for human race. If an AI won the next Hugo award, I would be rejoiced. It wouldn't mean the end of literature at all; it would mean that humans are ready t…

We already have a successful AI classical music writer https://en.m.wikipedia.org/wiki/Emily_Howell We have all kinds of visual art made by computers and AI - prom painting from photos to abstract art to 3D renders. We have computers writing poems and haiku. The only thing that's missing is the conceptual creation, which, let's be honest, most human artists struggle at as well. So writing and interesting story is not…

Art is more personal though. There is no single path to "winning" in art, and "good art" tends to mean different things to different people.

I'm sure soon (if not now) AI can easily create art that regurgitates popular trends in the past, and perhaps some artists may find a way to use AI / other algorithmic techniques in a way that complements their personal vision. But AI is a long way off from replicating the quirks of human nature, the unique personalities and personal visions of humans. Until that happens, I can't see terribly interesting art emerging from AI alone.

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

#394

Post by criddell » Thu, Mar 10, 2016, 4:36 PM UTC

TillE wrote: post #375

criddell wrote: post #287

Earlier quoted context omitted.

Literary awards are definitely subjective, but I don't think that matters. I'm just saying I think an AI will write a book that a significant portion of the population thinks is very good. The first "good" AI book will likely be from some formulaic genre (think 50 Shades of Gray ) that most people think is trash, but enough people will like it to legitimately propel it to the top of the sales charts. I also think my…

Music is a much easier problem (tight structure, lots of existing data to quickly analyze), and the animation bit is already being pretty thoroughly explored by procedural generation in games. You'd still need a "director" to pick shots, but most of the other pieces are nearly in place. We have algorithms that can create new environments, and design and animate new characters. But generating a coherent narrative, and…

> These are huge problems

Sure. I think they will be solved in the next 100 years though.

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

#395

Post by throwaway3301 » Thu, Mar 10, 2016, 4:36 PM UTC

pkaye wrote: post #339

Let's say AlphaGo can beat all the best human Go players. Then what will the next more difficult game for computers to compete against humans and win?

Poker remains unsolved (except for 2-player "Heads-up" limit hold'em[1]). Poker has many different variations and a player's strategy can change significantly based on many factors. The University of Alberta Computer Poker Research Group has already used counterfactual regret minimization in their Cepheus AI. I wonder if they could work with DeepMind to apply some of the techniques used in AlphaGo to poker.

[1]: http://science.sciencemag.org/content/347/6218/145

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

#396

Post by jdietrich » Thu, Mar 10, 2016, 4:36 PM UTC

fhe wrote: post #239

As someone who studied AI in college and am a reasonably good amateur player, I have been following the matches between Lee and AlphaGo. AlphaGo plays some unusual moves that go clearly against any classically trained Go players. Moves that simply don't quite fit into the current theories of Go playing, and the world's top players are struggling to explain what's the purpose/strategy behind them. I've been giving it…

The same thing happened in chess. Computers play in a very "computerish" way that was initially mocked, but became hugely influential on how humans play chess. Computer analysis opened up new approaches to the game.

http://www.nybooks.com/articles/2010/02/11/the-chess-master-...

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

#397

Post by statsaresimple » Thu, Mar 10, 2016, 4:37 PM UTC

astazangasta wrote: post #382

statsaresimple wrote: post #357

Earlier quoted context omitted.

As a follow up to your idea, we should explore two paths: first create the most powerful AI, second create subsystems devised to be interpretable. The powerful method could be used to train the interpretable method, that is we need an interpreter to translate from machine AI to human AI, and interpretable systems provide a middle ground.

I think training one function to approximate another function wouldn't help much; we'd inevitably lose the subtleties of the higher-order function and any insights that come with it. If we could train a decision tree to do what a CNN does and then interpret the outcome, why not use decision trees in the first place? I think the answer must be in figuring out how to decompose the black box of a CNN - it is, after all,…

Having a machine that gives you feedback in the middle of the game perhaps could be used to describe what is the weak point of a decision tree, and in which situations the method is good. It could detect some situations in which decision trees are good, then use that decision tree to understand what is happening and with that new understanding devise a new method in the middle. We could train a decision tree using new very powerful information about the value of the game in the middle of the game, that is new and powerful.

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

#398

Post by atom-morgan » Thu, Mar 10, 2016, 4:37 PM UTC

fhe wrote: post #239

As someone who studied AI in college and am a reasonably good amateur player, I have been following the matches between Lee and AlphaGo. AlphaGo plays some unusual moves that go clearly against any classically trained Go players. Moves that simply don't quite fit into the current theories of Go playing, and the world's top players are struggling to explain what's the purpose/strategy behind them. I've been giving it…

As a competitive speedcuber (Rubik's Cubes) this makes sense. If I watch a fellow cuber solve a cube, I understand their process even if it's a different method than the one I'd use. But a robot solving it? To my brain it looks like random turns until...oh shit it's finished.

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

#399

Post by ig1 » Thu, Mar 10, 2016, 4:38 PM UTC

1024core wrote: post #292

ig1 wrote: post #254

Earlier quoted context omitted.

AlphaGo is essentially built on the work that IBM did on TD-Gammon (a reinforcement learning backgammon player) in the 90s. Pretty much the same thing happened with TD-Gammon with it playing unconventional moves, in the longer term humans ended up adopting some of TD-Gammon's tactics once they understood how they played out, it wouldn't be surprising to see the same happen with Go.

> AlphaGo is essentially built on the work that IBM did on TD-Gammon (a reinforcement learning backgammon player) in the 90s. Citation needed.

And you'll find it in the AlphaGo paper. It's not a contentious claim.

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

#400

Post by botw » Thu, Mar 10, 2016, 4:40 PM UTC

EliRivers wrote: post #255

boucher wrote: post #246

Earlier quoted context omitted.

And yet the operative fact is that a human could not execute those calculations in a lifetime.

So what? Doesn't change the facts. There is no child here. There is a set of deterministic calculations written by some people, and executed.

please give the definition for a set of deterministic calculations.