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

AI

blog.samaltman.com

Post reply on HN
Print view View original

91–100 of 317 posts

Re: AI

#91

Post by cma »

mappum wrote: post #14

I believe human-level general intelligence (and beyond) is already inevitable, even if we don't make significant developments in "solving" intelligence. Projects that are already developing stuff like this (e.g. IBM Blue Brain) are just copying the human brain as closely as possible. Of course, this isn't as efficient as it could be (they simulate it all at the molecular level, so you can only get 1 neuron per CPU).…

They don't simulate it at the molecular level. It takes super computers like DE Shaw's Anton (hundreds of customized cores specifically designed for molecular simulation) weeks to simulate a few milliseconds of one single protein in water. IBM's approach is at a much higher level and is taking way more shortcuts.

Re: AI

#92

Post by marcosdumay »

edanm wrote: post #86

If anyone is interested in AI, I highly recommend joining Less Wrong, a community started by AI researcher Eliezer Yudkowsky. He started the community to convince people to focus on the "friendly AI problem". [1] I actually recommend that everyone read LW, but especially if you're interested in AI. [1] In a nutshell, the friendly AI problem is: assume we create an AI. It may rapidly become more intelligent than us, i…

> we'd better figure THAT out before building AI.

Why don't we just give that task to the AI? It'll be smarter than us...

Maybe the problem is that people are too easy to understand: We want "Brave New World", but we don't want to know about it, or that we want it.

Re: AI

#93

Post by Houshalter »

edanm wrote: post #86

If anyone is interested in AI, I highly recommend joining Less Wrong, a community started by AI researcher Eliezer Yudkowsky. He started the community to convince people to focus on the "friendly AI problem". [1] I actually recommend that everyone read LW, but especially if you're interested in AI. [1] In a nutshell, the friendly AI problem is: assume we create an AI. It may rapidly become more intelligent than us, i…

Defining human values completely and absolutely is an extremely difficult problem that might not even have a solution. AI on the other hand is probably inevitable sometime within the next century.

Re: AI

#94

Post by kordless »

I'm just going to stick this out there because I'm a futurist and it's my role to share with others what I'm thinking about. What I share may be flat out wrong, scary, half assed, or appear to be crazy. So be it.

We've advanced a lot in the last 100 years. We're starting to see a bigger picture forming with the advent of compute and networking capabilities. Combining simple elements of these basics give rise to surprising and interesting behaviors. See "Twitch Plays Pokemon": http://news.cnet.com/8301-1023_3-57619058-93/twitch-plays-po... as an example of suprising behavior.

The more we look in detail at the universe around us, the more puzzling it gets. Prime numbers spirals are unexplained. The two slit experiments results indicate the observer plays a part in collapsing a particle's probability wave. The effects of dark matter could be a result of parallel universes. You couldn't make up weirder shit if you tried.

It's not a huge leap of logic to assume some parts of our brain operate at a quantum level. Given that first statement comes to a truthful fruition, I don't think it would be entirely unreasonable to assume AI will do so as well. Given computers already use some quantum properties, it's also reasonable to expect advancement in AI lies in this direction.

When they announced Google was getting a D-Wave computer, I got really interested. Granted, they know beans about how it works (and whether or not it actually works at all) but it's still crazy interesting to consider.

As I said, I could be wrong or crazy. Or both.

Re: AI

#95

Post by Houshalter »

marcosdumay wrote: post #92
edanm wrote: post #86

If anyone is interested in AI, I highly recommend joining Less Wrong, a community started by AI researcher Eliezer Yudkowsky. He started the community to convince people to focus on the "friendly AI problem". [1] I actually recommend that everyone read LW, but especially if you're interested in AI. [1] In a nutshell, the friendly AI problem is: assume we create an AI. It may rapidly become more intelligent than us, i…

> we'd better figure THAT out before building AI. Why don't we just give that task to the AI? It'll be smarter than us... Maybe the problem is that people are too easy to understand: We want "Brave New World", but we don't want to know about it, or that we want it.

That's a viable approach to solving the problem. The question is, how do you trust the unfriendly AI to give you a correct solution?

Re: AI

#96

Post by benched »

Houshalter wrote: post #87
rectangletangle wrote: post #73

Earlier quoted context omitted.

But what is a problem that specifically requires intelligence to solve? How do we know we're measuring intelligence, and not the programs aptitude for the particular problem?

Why does it matter? If if solves the problem then it's as good as if it was fully intelligent. Who cares what goes on inside the black box? If you worried about it overfitting to a specific problem, give it lots of problems and weight the solutions by complexity. So you heavily favor simple algorithms that can learn to solve a large class of problems, over ones that are more adapted for those specific problems.

All I'm learning from this thread is that some people see infinite regression where others see tautology.

edit: what I mean:

Person 1: "What if we're just measuring the color of the sky? What if we're just measuring blueness? What if we're just measuring a wavelength of light? ... ..."

Person 2: "The sky looks blue."

Re: AI

#97

Post by carlob »

kordless wrote: post #94

I'm just going to stick this out there because I'm a futurist and it's my role to share with others what I'm thinking about. What I share may be flat out wrong, scary, half assed, or appear to be crazy. So be it. We've advanced a lot in the last 100 years. We're starting to see a bigger picture forming with the advent of compute and networking capabilities. Combining simple elements of these basics give rise to surpr…

> It's not a huge leap of logic to assume some parts of our brain operate at a quantum level.

To the best of our knowledge it is still a pretty large leap of logic.

Re: AI

#98

Post by benched »

rectangletangle wrote: post #67
seanalltogether wrote: post #59

How do we develop intelligent systems when we don't know how to measure intelligence?

This is a serious issue. The notion of "intelligence" is so abstract and intangible, that giving it a concrete definition is nearly redefining it. Likely what we're calling intelligence, is an oversimplification of reality; somewhat akin to the species problem in biology.

It is not universally believed that intelligence is ineffable. The most cursory Googling turns up lots of serviceable definitions articulated by smart people. Unfortunately, it is just complex enough to let people argue about it forever.

Re: AI

#99

Post by electrograv »

nl wrote: post #76
rdl wrote: post #36

Earlier quoted context omitted.

For me, "There are certainly some reasons to be optimistic. Andrew Ng, who worked or works on Google’s AI, has said that he believes learning comes from a single algorithm - the part of your brain that processes input from your ears is also capable of learning to process input from your eyes. If we can just figure out this one general-purpose algorithm, programs may be able to learn general-purpose things." is what w…

We already have a candidate algorithm[1], which is successful in a broad range of applications[2][3][4]. [1] http://en.wikipedia.org/wiki/Deep_learning#Convolutional_neu... [2] http://deeplearning.net/reading-list/ [3] http://en.wikipedia.org/wiki/Deep_learning#Results [4] http://www.wired.com/wiredscience/2012/06/google-x-neural-ne...

"Deep Learning" is not itself a candidate for anything, because it's not any single algorithm, but a category of approaches.

Deep Learning generally refers to machine learning algorithms that deal with stacking multiple layers of simpler functions to enable more complicated functions, and optimizing all the parameters to best fit your training set and generalize to new samples (the hard part). Though it usually refers to neural networks, I dont think there's any reason it doesn't also apply to other layered approaches as long as there's a relatively unified learning algorithm applied across the whole system.

There are clearly many different deep learning algorithms, even if you just count the permutations of tricks you can choose from to improve layered NN generalization. Though to be fair I think very good progress is being made towards developing "better" algorithms in the sense that new ones (e.g. RBM pretraining + dropout) usual perform better than than older algorithms, no matter what data you use it on (now network architecture is another matter entirely).

Re: AI

#100

Post by marcosdumay »

diminoten wrote: post #85
dicroce wrote: post #28

One thing that crosses my mind whenever I imagine creating a human level intelligence is that it takes humans YEARS of constant stimulation to begin to exhibit intelligent behavior... Sometimes I wonder if we'll have the algorithm may before we realize it...

We can stimulate a computer with the equivalent amount of information in much less time than years, though.

We probably can't. Unless it's an extremely fast algorithm, we don't have the processing power to make it run faster than our brain.

We'll can in a couple of decades, probably. But not now.

Post reply on HN