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

Artificial Intelligence Is Stuck

nytimes.com

Post reply on HN
Print view View original

81–90 of 189 posts

Re: Artificial Intelligence Is Stuck

#81

Post by giardini »

homarp wrote: post #71
lispm wrote: post #36

Earlier quoted context omitted.

The CERN budget is roughly a billion Euro per year.

some numbers from https://press.cern/facts-and-figures/budget-overview in 2016, 1127.2 million CHF or 1,163,675,662.22 USD

A billion dollars' research into AI would be more beneficial than another billion spent on CERN.

The Standard Model covers well over 99% of known physics already. Other moneys are being wasted paying students and professors to study string theory w/o any experiments possible.

Let's develop some true AI and let it close the gap. Two birds, one stone. [And maybe we can find out how we do analogies, at the same time!]

Re: Artificial Intelligence Is Stuck

#82

Post by kobeya »

bitL wrote: post #69

AI is currently about giving us an equivalent of a pocket knife for some tasks that are viewed as being in the "intelligence domain". That's all. Nobody really thinks we are anywhere close to general intelligence. Or like the famous saying "computers are bicycles for brain", current AI is about adding a small electric engine to those bicycles to make them easier for everyone.

> Nobody really thinks we are anywhere close to general intelligence

The kind of people that attend AGI conferences do.

Re: Artificial Intelligence Is Stuck

#83

Post by ethbro »

rwnspace wrote: post #10

I sense the hand of an editor. Particularly regarding the title. Embodiment seems to be a branch with low-hanging fruit, when it comes to advancing AGI. I think the economic structural problems are important, but it's possible to over-egg the details and for some lab to stumble on an experimental paradigm with features we didn't realise were implicated a priori. When it comes to other AIs, the idea that we are stuck…

The person you're largely agreeing with is Rodney Brooks (formerly of MIT CSAIL, http://people.csail.mit.edu/brooks/publications.html ).

And to sum it up a bit, the hypothesis is that humanlike AI is as much a product of the experience and reality of being physically (and limitedly!) human as it is any abstract algorithm.

You may also know him from a small company called iRobot (aka Roomba).

Re: Artificial Intelligence Is Stuck

#84

Post by kobeya »

mannykannot wrote: post #65
bhickey wrote: post #58

Earlier quoted context omitted.

I'm riffing on his writing -- > Even Google Translate, which pulls off the neat trick of approximating translations by statistically associating sentences across languages, doesn’t understand a word of what it is translating. This is just another incarnation of "AI is the thing we haven't done." He's parroting Chomsky's disdain for statistical models and John Searle's fundamental misunderstanding of AI. For the forme…

>> Even Google Translate, which pulls off the neat trick of approximating translations by statistically associating sentences across languages, doesn’t understand a word of what it is translating. >This is just another incarnation of "AI is the thing we haven't done." I don't think so - it appears to be an objectively correct assessment of the current state of the art. > Otherwise we're just moving goalposts. The fir…

> On the other hand, I am surprised by the claim that AI is stuck; my outsider's impression is that progress has accelerated. Perhaps the impression of being stuck comes from more people realizing how difficult a problem it is.

Deep learning made practical a large number of applications that were previously intractable by neural network approaches. Advances over the last 10 years have pushed the boundaries of what machine learning systems are capable of doing. However machine learning has algorithmic limits to what it can accomplish, and we are starting to hit those limits. A change in paradigm is required to begin making real progress again. Either a change to something new or a regression to older ideas that were temporarily put on the back burner.

That's not a universally held view, but I think it is the sentiment behind this editorialized title.

Re: Artificial Intelligence Is Stuck

#85

Post by cscurmudgeon »

tim333 wrote: post #74

I'd dispute that AI research is stuck or that his proposed answer along the lines of "An international A.I. mission focused on teaching machines to read" is a good one. Research seems to be cracking along with AlphaGo, self driving cars and the like. Recently DeepMind have been doing interesting stuff with dreams[1], imagination[2] and body movement[3], the last one being a little reminiscent of his daughter inventin…

See https://news.ycombinator.com/item?id=14880853 for one data point on why research is stuck.

Re: Artificial Intelligence Is Stuck

#86

Post by giardini »

scardine wrote: post #37
asketak wrote: post #9

Most of the AI progress in last years is just tuning pattern recognition algorithms. We can not expect these algorithms to produce results like humans, because humans have a lot of information not just from percieving the world, but their patterns of thinking are also vastly dependend on the underlying structure of brain, that has developed over milions of years of evolution. If there is a cliff, toddlers are scared…

You must watch the documentary "The secret life of babies". There is a scene where they put several babies on a "cliff" situation (with a glass in the same level to prevent them falling), they will confidently march into the glass - they do several other experiments that prove babies are fearless. https://www.netflix.com/title/80009352

Babies are not born afraid of the cliff (i.e., it isn't hardwired into the infant brain); they develop fear of it at about 9 months' age, once their depth perception has developed.

Re: Artificial Intelligence Is Stuck

#87

Post by visarga »

rwnspace wrote: post #10

I sense the hand of an editor. Particularly regarding the title. Embodiment seems to be a branch with low-hanging fruit, when it comes to advancing AGI. I think the economic structural problems are important, but it's possible to over-egg the details and for some lab to stumble on an experimental paradigm with features we didn't realise were implicated a priori. When it comes to other AIs, the idea that we are stuck…

> Embodiment seems to be a branch with low hanging fruit, when it comes to advancing AGI.

If it were "low hanging" it would have been picked already. Reinforcement learning with AI agents is hard, especially in a dynamic environment with many types of objects.

I think the path towards AGI is to do simulation coupled with deep learning. Simulation would open the door to predicting non-trivial effects that cannot be learned by example because they are so rare that there are no training examples. We can generate artificial training examples to cover all the rare cases.

Re: Artificial Intelligence Is Stuck

#88

Post by ethbro »

fauigerzigerk wrote: post #31

I can't follow the author's logic. First he complains about the limited breadth of AI approaches (bottom up) and then he makes the case for more central coordination of research efforts. Contrary to applied physics or medicine, AI doesn't require massive capital investment like building a particle accelerator or running clinical trials over years. So if we already suffer from a lack of diversity, why should we ape th…

The author's general problem is that he seems ignorant (or willfully ommiting) the expert systems period in AI (realized and popular in the 80s, academic foundations discovered in the 60s and 70s).

I agree with the article that GP AI is likely to ultimately be a fusion of bottom-up with top-down systems, and that expert systems seem to be getting short shift after their earlier failures while neural networks are possibly receiving overly optimistic expections.

To be fair, I believe this is the author: https://en.m.wikipedia.org/wiki/Gary_Marcus , and he appears to have a cognitive neuroscience background as opposed to computational AI. So I wouldn't be surprised if he actually was unaware of 1960s-80s CS AI research.

Re: Artificial Intelligence Is Stuck

#89

Post by giardini »

nl wrote: post #35
scotty79 wrote: post #30

Earlier quoted context omitted.

https://youtu.be/WanGt1G6ScA Not sure what to think about it.

https://youtu.be/WanGt1G6ScA?t=92 (from that video, at a timepoint which shows a young a baby crawling over a glass surface with a drop beneath it, with no sign of discomfort). I'd say this proves the point completely. It shows that babies aren't scared of heights, and they don't acquire the fear from experience. Instead it seems likely it is that they either develop a mental model of how the world works, or they lea…

Or the mechanism is developed during a "critical period" once depth perception has developed. It isn't necessarily totally learned behavior.

https://en.wikipedia.org/wiki/Critical_period

Re: Artificial Intelligence Is Stuck

#90

Post by rwnspace »

ethbro wrote: post #83
rwnspace wrote: post #10

I sense the hand of an editor. Particularly regarding the title. Embodiment seems to be a branch with low-hanging fruit, when it comes to advancing AGI. I think the economic structural problems are important, but it's possible to over-egg the details and for some lab to stumble on an experimental paradigm with features we didn't realise were implicated a priori. When it comes to other AIs, the idea that we are stuck…

The person you're largely agreeing with is Rodney Brooks (formerly of MIT CSAIL, http://people.csail.mit.edu/brooks/publications.html ). And to sum it up a bit, the hypothesis is that humanlike AI is as much a product of the experience and reality of being physically (and limitedly!) human as it is any abstract algorithm. You may also know him from a small company called iRobot (aka Roomba).

Thank you very much for the link, and nice summary. This looks very interesting, I'm grateful: my learning habits mean I'm liable to playing with second-hand scraps of ideas, and missing out on sources.
Post reply on HN