Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

What AlphaGo Zero teaches us about what’s going wrong with innovation

timharford.com

Post reply on HN
Print view View original

101–110 of 116 posts

Re: What AlphaGo Zero teaches us about what’s going wrong with innovation

#101

Post by cosmic_ape »

Not only corporations abandon real fundamental research, but people like the the author of that blog start referencing things like AlphaGo as "fundamental research" comparable to Shannon's or Turing's work.

AlphaGo is a good and necessary engineering, but the ideas are pretty old, and not especially illuminating. Start confusing it with fundamental research often enough, and people will start to believe it. And then, corporate managers, and academic research grants, and academic publishing venues, like conferences, will start expecting "fundamental research" to be like AlphaGo instead of actual fundamental research.

Re: What AlphaGo Zero teaches us about what’s going wrong with innovation

#102

Post by wallflower »

nine_k wrote: post #42
wallflower wrote: post #34

Earlier quoted context omitted.

If you read the science fiction series "The Three-Body Problem" (highly recommended), it makes a very compelling argument that fundamental research is the most important investment in the future. For example, fusion drives, not traditional stored rocket propellant engines, will be necessary to navigate between planets and the outer solar system. Also, existing known behaviors/laws of physics aside, the book posits th…

The mother planet reaps no material benefits from colonizing another star. Those groups who colonize it get a world if their own. So interests of Earth governments are not aligned with star travel, and only marginally aligned with colonizing e.g. Mars. Those groups who want to ho there will have to do it themselves. (See Elon Musk.)

> The mother planet reaps no material benefits from colonizing another star.

In the book series, without giving away too much of the plot, the handful of countries that acquire and develop spaceflight technology/space bases become political powers in their own right.

If you watch or read "The Expanse", the political shifts are similar. The colony on Mars, in particular, is particularly threatening and powerful.

Re: What AlphaGo Zero teaches us about what’s going wrong with innovation

#103

Post by forgot-my-pw »

Join Leela Zero in trying to replicate AlphaGo Zero: https://github.com/gcp/leela-zero

It's estimated that AlphaGo Zero took about 1700 GPU years to train. We can only reach this number by having a distributed effort.

400k games has currently been submitted to the Leela Zero project: http://zero.sjeng.org/ . It's still playing in amateur level. (AGZ had about 30m self-play training).

Re: What AlphaGo Zero teaches us about what’s going wrong with innovation

#104

Post by dwaltrip »

JustFinishedBSG wrote: post #41
dsacco wrote: post #38

Earlier quoted context omitted.

While I understand your cynicism in the practical applicability of a chess or go-playing AI, I think you are significantly underestimating the theoretical innovations contributed to the field every time these models are substantially improved. Much of the work that goes into improving something like AlphaGo is cross-applicable and cross-pollinated to other research projects, and gradually trickles out into other doma…

> I think you are significantly underestimating the theoretical innovations contributed to the field every time these models are substantially improved. I think you are overestimating, there isn't a single interesting theoretical insight in AlphaGo's papers.

That is a big claim. Where's the detailed analysis and citations?

Re: What AlphaGo Zero teaches us about what’s going wrong with innovation

#105

Post by fjsolwmv »

ionforce wrote: post #83
lustig wrote: post #44

Earlier quoted context omitted.

Technically yes, most often it's about stacking more layers in neural networks, making them "deep". However, there is some merit to the new hype since stacking more layers worked way better than anyone previously working with neural networks and ML thought it would. But in theory you could generalize deep learning to other methods than neural networks, it's basically about creating way more complex models than those…

I thought all neural networks had layers. Is this not the case?

Deep networks have more layers than the previous generation. They seem to work better in engineering practice than mathematically equivalent short wide networks.

Re: What AlphaGo Zero teaches us about what’s going wrong with innovation

#106

Post by fjsolwmv »

icc97 wrote: post #29
dingo_bat wrote: post #16

I don't think there is a very sharp distinction between results oriented R&D and "basic research". In the article, IBM's deep blue is dismissed as a dead-end victory but apparently alphago is not? Why? They both seem identical to me in goals and research methodology. On a side note, I cannot wait for general super intelligence. It cannot come soon enough. I'm tired of being poor and stuck in a fucking rut, and contem…

I think the point was that Deep Blue was using brute force, where as Alpha Go had taught itself and Alpha Go Zero has gone the full distance to require zero outside help. So Deep Blue wasn't such a big step in terms of General Artificial Intelligence because it was heavily dependent on the human optimisations and was just showing the power of number crunching rather than learning.

It's more of a continuum than that. Yes deep blue had more hand-tuning and custom-built structure. AlphaGo had less hand tuning and of course waaaay more brute Force on top.

Re: What AlphaGo Zero teaches us about what’s going wrong with innovation

#107

Post by fjsolwmv »

dagw wrote: post #20
dingo_bat wrote: post #16

I don't think there is a very sharp distinction between results oriented R&D and "basic research". In the article, IBM's deep blue is dismissed as a dead-end victory but apparently alphago is not? Why? They both seem identical to me in goals and research methodology. On a side note, I cannot wait for general super intelligence. It cannot come soon enough. I'm tired of being poor and stuck in a fucking rut, and contem…

Why? They both seem identical to me in goals and research methodology. In theory taking the work done on AlphaGo (and more importantly AlphaGo Zero) and generalizing it to non-Go related problems should be a lot easier than taking the work done on deep blue and generalizing to non chess related problem.

AlphaGo is an application of general tech, just like Deep Blue was an application of general tech.

Re: What AlphaGo Zero teaches us about what’s going wrong with innovation

#108

Post by fjsolwmv »

teabee89 wrote: post #68

I highly recommend this talk "Greatness cannot be planned: the Myth of the Objective" by Kenneth Stanley: he created picbreeder.org (evolutionary art platform) and realized that if an interesting state is set as an objective, then it is extremely hard to reach it from the initial state with AI algorithms, because you need to move away sometimes a lot, from local optima.

The notion of local optima is centuries older than the invention of the first .org

Re: What AlphaGo Zero teaches us about what’s going wrong with innovation

#109

Post by argonaut »

mda wrote: post #45
GuB-42 wrote: post #35

Projects like DeepBlue and AlphaGo are not fundamental innovation nor research, they are just PR stunts that show the expertise of the company making them. TBH, winning a game of chess or go has little value in itself, except for the limited market of selling chess or go software. The reason they are doing that is mostly for publicity. IBM makes computers, and they show how good they are at it by having one beat top…

Funny that up until 2016 go was regarded as one of the most difficult games that computers could master, and now that it is solved it becomes a PR stunt? Would you claim the same in 2015?

This is one of those catchy but actually unverified memes that gets endlessly propagated.

"Nevertheless, I believe that a world-champion-level Go machine can be built within 10 years" - Feng-Hsiung Hsu, 2007 (researcher who worked on Deep Blue). https://spectrum.ieee.org/computing/software/cracking-go

Re: What AlphaGo Zero teaches us about what’s going wrong with innovation

#110

Post by klmr »

patall wrote: post #97
klmr wrote: post #53

Earlier quoted context omitted.

> Medical research has advanced a lot with the invention of CRISPR. You are overestimating the importance of CRISPR. Don’t get me wrong — it is hugely important and innovative. But it’s not even the most important biomedical research innovation of recent years (I’d give that title to RNA-seq or more generally next-gen sequencing technology; but there are several contenders). But at any rate all the innovations you li…

I beg to differ, it is hard to overestimate the importance of CRISPR. There have been gene editors like zink-fingers or TALENs before but CRISPR is in its total capability is a true breakthrough. NGS on the other side is more or less a gradual development. While the exact technology may be novel or unique the whole process is not as can be seen by the various competing technologies that existed and still do exist wit…

> NGS on the other side is more or less a gradual development.

While true that’s not really relevant. What’s relevant is that it has completely revolutionised biomedical research. And although CRISPR has the potential of doing the same, it’s just telling that RNA-seq (and related technologies) have become so routine that they’ve effectively spawned new fields of research. Together with WGS (and preceded by microarrays), the new sequencing technologies have led to a revolution of how science is done: because most of the science here is done at a computer. CRISPR, by contrast, is “merely” a new molecular biology tool. A very powerful, for sure, and one that opens up completely new avenues of research. But it doesn’t fundamentally change the way we do science. NGS has.

Post reply on HN