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

What you wanted to know about AI

fastml.com

Post reply on HN
Print view View original

81–90 of 165 posts

Re: What you wanted to know about AI

#81

Post by therealwill »

vlunkr wrote: post #68
Retric wrote: post #62

Earlier quoted context omitted.

I suspect that the current google car is already safer than the overall average driver. Don't forget some people speed, flee from the cops, fall asleep at the wheel, get drunk, text, look at maps, have strokes etc. So, sure at peak performance cars have a long way to go. However, accidents are often a worst case and computers are very good at paying attention to boring things for long periods of time. PS: If driverle…

You might be right that they are safer, but you're totally failing to contest the point. Yes computers are lots better than humans at lots of things, but they are worse at other things. They can't reliably tell the difference between a cat and a bench. Things that may be important when the computer is going 80 mph with humans on board.

Why would a self driving car need to know the difference between a cat and a bench? All it really has to know is that it is an object of a certain size and not to hit it.

The things that the car needs to know are largely standardized: the lines on the road, road signs, speed limits, etc.

Re: What you wanted to know about AI

#82

Post by rwallace »

one-more-minute wrote: post #70
rwallace wrote: post #46

Earlier quoted context omitted.

If people had started to worry about the long-term effects of carbon output before it was already widespread, the Industrial Revolution would have been strangled at birth (the road from wood fires to solar panels leads through coal-burning steam engines; refuse to ever burn coal in large quantities and the chain is broken), and we would have stumbled around banging the rocks together until evolution optimized general…

Who said anything about refusing to burn coal in large quantities? The benefits of the industrial revolution may well have outweighed the costs, and no one is saying it was a bad thing – or likewise, that developments in AI are a bad thing. But are you saying there's absolutely nothing that could've been done better in the industrial revolution? With some foresight we might have thought to develop solar power more ur…

I'm saying there is nothing that would in fact have been done better by premature worry about global warming. If your argument is that any sequence of actions will be less than optimal - compared to, say, the actions that could have been carried out by a hypothetical omniscient entity of infinite wisdom and benevolence - then that is certainly true but irrelevant. But the previous argument was that convincing actual humans in the nineteenth century to start worrying about global warming would have been net beneficial, and I'm pointing out it would have been disastrous.

And that, mind you, is still with the anachronistic application of 21st-century knowledge to the matter. If people in the 19th century had actually tried to figure out what they should be worrying about if they were going to worry two centuries prematurely, they would likely have come up with something completely different.

Re: What you wanted to know about AI

#83

Post by FiatLuxDave »

jallmann wrote: post #39

When I started school, my dream was to figure out a theory to underpin a grand unified model of artificial intelligence. Imagine my disappointment once I started studying the subject in detail. Most functional AI nowadays consists of algorithms that are carefully tuned to solve a very specific problem in a narrowly defined environment. All the research nowadays is pushing the boundaries of a local optimum. Right now,…

There is a problem with naming. There are two kinds of AI which get discussed - AI (Artificial Intelligence) and AGI (Artificial General Intelligence).

The problem with these names are that they are not easily distinguishable by the general public or others who are not already familiar with the concepts. Its like the poor coding pattern where you have two variables in your code with almost the same names - 'wfieldt' and 'wtfield' will get confused by the next person who has to maintain your code. It's the same problem with popularization of concepts.

The same thing happens in Physics, where people confuse energy with power, and think that Special Relativity must be much harder than General Relativity (it's special, right?). The words used to name things make them understandable, or not.

What's worse is that this confusion is intentional, created on the part of people who wanted to hype their (actually valuable, but perhaps overlooked) work, by confusing their 'boring' or 'limited' AI work with AGI work in the minds of those supporting or funding that work.

That's why people say things like "true AI".

There really needs to be a name for the non-AGI kind of AI, which refers specifically to the limited-domain type of problem solving which "AI" has been so successful. "AI" is a bad name for this.

Personally, I am partial to the metaphor of "canned thought". It shows that a human had to do the work beforehand of setting up the structure of the algorithm that is later doing the information processing, and that the process is limited by the thought that were previously put into it. But Canned Thought is also a pretty bad name for the field. It's not descriptive and it's not catchy. Anyone have any suggestions for a better name for non-AGI AI?

Re: What you wanted to know about AI

#84

Post by themgt »

jallmann wrote: post #39

When I started school, my dream was to figure out a theory to underpin a grand unified model of artificial intelligence. Imagine my disappointment once I started studying the subject in detail. Most functional AI nowadays consists of algorithms that are carefully tuned to solve a very specific problem in a narrowly defined environment. All the research nowadays is pushing the boundaries of a local optimum. Right now,…

The core dilemma, and I'm paraphrasing I can't remember who, is that for every other problem given to computer scientists to solve, there already exists an understanding of what the problem space is and what a solution is required to do. The computer scientist just architects and implements the solution in software, but they're implementing business logic, or physics equations to guide a spacecraft, or "route this car through the mapped 3d environment and follow these rules and don't hit anything"

But sentience simply is not defined. There are no equations for it, no foundational science modeling it. It's possibly the ultimate unanswered question for all human history, and so the task given to computer engineers is: implement this thing we haven't as a species been able to explain or describe or comprehend for 10,000 years

And really, it's a bit much to ask. The neuroscientists could have a eureka moment tomorrow and finally crack the code of sentience, and once documented I'm confident some form of rudimentary synthetic implementation would be possible within a decade. But I'm not sure many neuroscientists or practically anyone (Tononi/Tegmark/Hawkins/?) are directly attempt to investigate and build theories of consciousness.

That theory of consciousness would be the instruction book for the programmers building the first hard AI.

Re: What you wanted to know about AI

#85

Post by AndrewKemendo »

maaku wrote: post #61
AndrewKemendo wrote: post #50

Earlier quoted context omitted.

at this point there's a respectable literature on the potential dangers of unconstrained, powerful optimizing agents, of which I linked a couple of examples. There isn't really though. Superintelligence, while an interesting book, doesn't do much more than pontificate on sci-fi futures based largely on writing from Yudkowski. Bostrom gives no clear pathway to AGI. Neither does Yudkowski in his own writings or Barrat…

I'm more in your camp than MIRI's but... Could you argue against the position directly? You kinda took a swerve in the middle there. Bostrom, Yudkowsky, and Barrat's positions are basically "superintelligence is possible, it will eventually happen in some way, and when it does ..." It's not within the scope of the quoted works to elaborate a technical roadmap or provide firm dates for the arrival of superhuman machin…

Could you argue against the position directly?

No I can't because there isn't a way to argue it. I can't argue that it would be safe because I have nothing to point to that says it will be or even could be. We simply do not know enough about how it would be built to reason on it.

It's not within the scope of the quoted works to elaborate a technical roadmap or provide firm dates for the arrival of superhuman machine intelligence.

Correct, but they don't even reference a technical road map because there isn't one. It's too early to tell.

Do you have any points regarding this particular difference of opinion?

Not really, other than my own opinions which are as baseless as the others. I mean I have my own opinions and thoughts on things but they aren't empirically based at all, which is my main contention. We can't reason, let alone start making policy on stuff that we haven't the slightest idea how to build.

Re: What you wanted to know about AI

#86

Post by AndrewKemendo »

FiatLuxDave wrote: post #83
jallmann wrote: post #39

When I started school, my dream was to figure out a theory to underpin a grand unified model of artificial intelligence. Imagine my disappointment once I started studying the subject in detail. Most functional AI nowadays consists of algorithms that are carefully tuned to solve a very specific problem in a narrowly defined environment. All the research nowadays is pushing the boundaries of a local optimum. Right now,…

There is a problem with naming. There are two kinds of AI which get discussed - AI (Artificial Intelligence) and AGI (Artificial General Intelligence). The problem with these names are that they are not easily distinguishable by the general public or others who are not already familiar with the concepts. Its like the poor coding pattern where you have two variables in your code with almost the same names - 'wfieldt'…

There really needs to be a name for the non-AGI kind of AI, which refers specifically to the limited-domain type of problem solving which "AI" has been so successful. "AI" is a bad name for this.

I think you are looking for Narrow Artificial Intelligence, which hasn't gained much traction. To me Narrow and General AI respectively make perfect sense for their scopes.

[1]http://en.wikipedia.org/wiki/Weak_AI

Re: What you wanted to know about AI

#87

Post by im3w1l »

AndrewKemendo wrote: post #50
davmre wrote: post #45

Earlier quoted context omitted.

I'm not sure I understand what you mean by "straw man" here. The usual meaning is that to attack a straw man means to argue against a position that no one actually holds, but which is easier to attack than your opponent's actual position. The concerns about the long-term future of AI are real, actual beliefs held by serious people. At this point there's a respectable literature on the potential dangers of unconstrain…

at this point there's a respectable literature on the potential dangers of unconstrained, powerful optimizing agents, of which I linked a couple of examples. There isn't really though. Superintelligence, while an interesting book, doesn't do much more than pontificate on sci-fi futures based largely on writing from Yudkowski. Bostrom gives no clear pathway to AGI. Neither does Yudkowski in his own writings or Barrat…

>For example, find me someone who can tell me what kind of hardware we need for an AGI.

My guess: Something with tons of flops (or even fixed point ops, if that would be easier). Everything else secondary.

Re: What you wanted to know about AI

#88

Post by leeoniya »

jallmann wrote: post #39

When I started school, my dream was to figure out a theory to underpin a grand unified model of artificial intelligence. Imagine my disappointment once I started studying the subject in detail. Most functional AI nowadays consists of algorithms that are carefully tuned to solve a very specific problem in a narrowly defined environment. All the research nowadays is pushing the boundaries of a local optimum. Right now,…

surprised not to see this linked in the article:

http://karpathy.github.io/2012/10/22/state-of-computer-visio...

Re: What you wanted to know about AI

#89

Post by AndrewKemendo »

cousin_it wrote: post #73
AndrewKemendo wrote: post #29

Earlier quoted context omitted.

How about this instead: Current AI technologies give us no insight about the probability of an AGI Because that is exactly the point. We can't look at what we have built today and extrapolate that AGI will come from it. All anyone can do is speculate wildly on AGI just as they were able to in 1956 [1]. The difference is that we see automation/computing as ubiquitous (smartphones/computers in 2015), rather than around…

Probability of AGI is close to 100%, conditional on mankind surviving long enough. We know that human brains don't run on magic, therefore you can build a machine that can do the same thing. (Turing computability doesn't matter, if the brain relies on non-Turing-computable physical effects then a machine can rely on those as well.) The current situation where unmodified humans are the smartest creatures around is uns…

We know that human brains don't run on magic, therefore you can build a machine that can do the same thing.

I mean that is what I think, and most of the community thinks, but we haven't proven that empirically - nor could we, its unprovable.

But you don't have any good arguments for feeling safe.

Correct. Notice that I am not saying that we are safe. I personally think we aren't safe [1], but that is a different discussion which is equally as baseless.

The key here is that the argument that AGI will not be safe, and therefore needs to be regulated to make it safe, not only misses the point but ensures that we would never make one. The current level of funding for AGI development and understanding of the roadmap for AGI is absolutely dismal given it's potential impact. If we now add the constraint that in order to make one, we also ensure it doesn't threaten humanity - then you have sounded the death knell for AGI. Once we make AGI - in the sense that it is a recursively self improving system with human equivalent capabilities across domains - there is no way we will know what will happen. That's it.

[1] Which is why I am personally in the transhumanist camp.

Re: What you wanted to know about AI

#90

Post by aamar »

AndrewKemendo wrote: post #40
Thrymr wrote: post #35

Earlier quoted context omitted.

Arrhenius predicted global warming in 1896: http://en.wikipedia.org/wiki/Svante_Arrhenius#Greenhouse_eff...

Ah, Fantastic! He led the creation of the field of climatology if I recall. If someone can do forecasting of AGI with similar empiricism that would be revolutionary and welcomed.

Arrhenius probably did no empirical research in this field. He devised a generalized formula based on theoretical considerations and indirect measurements (of the moon's appearance). It would take ~60 years before a quorum of researchers believed his formula could provide accurate real-life predictions.

Moravec's 70s papers are roughly comparable to Arrhenius's formula. The basic computational power predictions (a minor extension of Moore's law) still seem to be basically correct. There's a reasonable debate to be had about whether more interesting predictions made by Moravec/Kurzweil/etc. have been validated or not.

Post reply on HN